Building a Knowledge Base for Global Action

Home , 1993 in Uganda, 1998 in Sudan, Acraea acerata, Bakia, Pepper leaf curl virus, Tomatillo

About the Partners

CIAT The International Center for Tropical Agriculture (CIAT) is a not-for-profit organization that conducts socially and environmentally progressive research aimed at reducing hunger and poverty and preserving natural resources in developing countries. CIAT is one of the 15 centers funded mainly by the 58 countries, private foundations, and international organizations that make up the Consultative Group on International Agricultural Research (CGIAR).

IITA IITA’s mission is to enhance the food security, income, and well-being of resource-poor people in sub-Saharan Africa by conducting research and related activities to increase agricultural production, improve food systems, and sustainably manage natural resources, in partnership with national and international stakeholders.

AVRDC AVRDC is the principal international center for vegetable research and development in the world. Its mission is to reduce poverty and malnutrition in developing countries through improved production and consumption of vegetables.

CIP The International Potato Center (CIP) seeks to reduce poverty and achieve food security on a sustained basis in developing countries through scientific research and related activities on potato, sweetpotato, and other root and tuber crops and on the improved management of natural resources in the Andes and other mountain areas.

ICIPE ICIPE’s mission is to help alleviate poverty, ensure food security and improve the overall health status of peoples of the tropics by developing and extending management tools and strategies for harmful and useful arthropods, while preserving the natural resource base through research and capacity building.

About the Donors

DANIDA Reducing poverty in developing countries is central to Danish development cooperation priorities. A number of crosscutting themes are built into DANIDA’s development assistance: women’s participation in development, the environment, promotion of democracy and observation of human rights. These crosscutting themes are integrated into DANIDA’s development activities more generally.

DFID The United Kingdom’s Department for International Development (DFID) supports research in the sustainable management of natural resources that aims at improving the livelihoods of poor people in developing countries. DFID has funded commodity-based research on whiteflies and whitefly- transmitted viruses, primarily in Africa, for nearly a decade through its centrally funded Crop Protection Programme (CPP). It is now aiming to build on this support by helping to create a global framework that will allow different whitefly research projects and activities to coordinate their efforts more effectively and produce research outputs which have positive and sustainable impacts on poor peoples’ livelihoods.

NZAID NZAID is New Zealand’s agency responsible for international assistance to developing countries. New Zealand’s aid helps to eliminate poverty and create a safe and just world, particularly in the Pacific region.

USAID USAID is an independent federal government agency that receives overall foreign policy guidance from the Secretary of State. USAID supports long-term and equitable economic growth and advances U.S. foreign policy objectives by supporting: economic growth, agriculture and trade; global health; and, democracy, conflict prevention and humanitarian assistance.

USDA Provides leadership on food, agriculture, natural resources, and related issues based on sound public policy, the best available science, and efficient management. ISBN 958-694-074-8

Whitefly and Whitefly-borne Viruses in the Tropics: Building a Knowledge Base for Global Action

Edited by: Pamela K. Anderson and Francisco J. Morales

With the collaboration of Annie L. Jones and Richard H. Markham Centro Internacional de Agricultura Tropical International Center for Tropical Agriculture Apartado Aéreo 6713 Cali, Colombia Fax: +57 (2) 4450073 E-mail: [email protected]

CIAT Publication No. 341 ISBN 958-694-074-8 Press run: 500 Printed in Colombia August 2005

Whitefly and whitefly-borne viruses in the tropics: Building a knowledge base for global action / edited by Pamela K. Anderson and Francisco J. Morales; with the collaboration of Annie L. Jones and Richard H. Markham. Cali, CO : Centro Internacional de Agricultura Tropical (CIAT), 2005. 351 p. -- (CIAT publication no. 341) ISBN 958-694-074-8

AGROVOC descriptors in English: 1. Manihot esculenta. 2. Ipomoea batatas. 3. Sweet potatoes. 4. Vegetables. 5. Legumes. 6. Pest insects. 7. Aleyrodidae. 8. Bemisia tabaci. 9. Trialeurodes vaporariorum. 10. Aleurotrachelus socialis. 11. Trialeurodes variabilis. 12. Bemisia tuberculata. 13. Trialeurodes abutiloneus. 14. Aleurocanthus woglumi. 15. Cassava mosaic virus. 16. Geminiviruses. 17. Vectors. 18. Food security. 19. Poverty. 20. Pest control. 21. Disease control. 22. Biological control. 23. Integrated control. 24. Parasitoids. 25. Natural enemies. 26. Resistance to chemicals. 27. Socioeconomic environment. 28. PCR. 29. RAPD. 30. Mixed cropping. 31. Highlands. 32. Capacity building. 33. Africa. 34. Central America. 35. Mexico. 36. Caribbean. 37. Latin America.

Local descriptors in English: 1. Cassava. 2. Whitefly. 3. Molecular techniques.

AGROVOC descriptors in Spanish: 1. Manihot esculenta. 2. Ipomoea batatas. 3. Batatas. 4. Hortalizas. 5. Leguminosas. 6. Insectos dañinos. 7. Aleyrodidae. 8. Bemisia tabaci. 9. Trialeurodes vaporariorum. 10. Aleurotrachelus socialis. 11. Trialeurodes variabilis. 12. Bemisia tuberculata. 13. Trialeurodes abutiloneus. 14. Aleurocanthus woglumi. 15. Virus del mosaico de la cassava. 16. Geminivirus. 17. Vectores. 18. Seguridad alimentaria. 19. Pobreza. 20. Control de plagas. 21. Control de enfermedades. 22. Control biológico. 23. Control integrado. 24. Parasitoides. 25. Enemigos naturales. 26. Resistencia a productos químicos. 27. Entorno socioeconómico. 28. PCR. 29. RAPD. 30. Cultivo mixto. 31. Zona de montaña. 32. Creación de capacidad. 33. África. 34. América Central. 35. México. 36. Caribe. 37. América Latina.

Local descriptors in Spanish: 1. Yuca. 2. Mosca blanca. 3. Técnicas moleculares.

I. Anderson, Pamela K. II. Morales, Francisco José. III. Centro Internacional de Agricultura Tropical. IV. Ser.

AGRIS subject categories: H10 Pests of plants / Plagas de plantas H20 Plant diseases / Enfermedades de las plantas

LC classification: SB 945 .W4 W4

CIAT encourages wide dissemination of its printed and electronic publications for maximum public benefit. Thus, in most cases colleagues working in research and development should feel free to use CIAT materials for noncommercial purposes. However, the Center prohibits modification of these materials, and we expect to receive due credit. Though CIAT prepares its publications with considerable care, the Center does not guarantee their accuracy and completeness. Contents

Page Foreword vii Introduction Pamela K. Anderson 1

SECTION ONE Whiteflies as Vectors of Plant Viruses in Cassava and Sweetpotato in Africa Chapter 1.1 Introduction James Legg 15 1.2 Ghana Anthony Cudjoe, Joseph Gyamenah and Braima James 24 1.3 Benin Brice Gbaguidi, Braima James and Symphorien Saizonou 30 1.4 Nigeria Nnamdi C. Echendu, Joseph B. Ojo, Braima James and Brice Gbaguidi 35 1.5 Cameroon Nelson Ntonifor, Braima James, Brice Gbaguidi and Ambe Tumanteh 40 1.6 Uganda Peter Sseruwagi, James Legg and William Otim-Nape 46 1.7 Kenya Joseph Kamau, Peter Sseruwagi and Valente Aritua 54 1.8 Tanzania Joseph Ndunguru, Simon Jeremiah, Regina Kapinga and Peter Sseruwagi 61

iii Whiteflies and Whitefly-borne Viruses in the Tropics

Page Chapter 1.9 Malawi Mathews P. K. J. Theu and Peter Sseruwagi 68 1.10 Madagascar Sahondramalala Ranomenjanahary, Jocelyn Ramelison and Peter Sseruwagi 72 1.11 The Diversity of Cassava Mosaic Begomoviruses in Africa Peter Markham, Robert Briddon, Clotilde Roussot, John Farquhar, Geoffrey Okao-Okuja and James Legg 77 1.12 Serological Analysis of Sweetpotatoes Affected by Sweetpotato Virus Disease in East Africa Valente Aritua, Richard Gibson and Josef Vetten 83 1.13 Factors Associated with Damage to Sweetpotato Crops by Sweetpotato Virus Disease Richard Gibson, Valente Aritua and Simon Jeremiah 89 1.14 Conclusions and Recommendations James Legg and Braima James 98

SECTION TWO Whiteflies as Pests and Vectors of Viruses in Vegetable and Legume Mixed Cropping Systems in Eastern and Southern Africa 2.1 Introduction Mohamed Ali Bob, Rebecca Raini and Bernhard Löhr 115 2.2 Sudan Gassim Dafalla and Musa Ahmed 118 2.3 Kenya Mohamed Ali Bob, Benjamin Odhiambo, Gilbert Kibata and Jason Ong’aro 129 2.4 Tanzania Ignas Swai and Simon Slumpa 141 2.5 Malawi Harriet Thindwa and Patric Khonje 150 2.6 Tomato Yellow Leaf Curl Virus and Tomato Leaf Curl-like Viruses in Eastern and Southern Africa Remi Nono-Womdim, Ignas Swai, Madan Mohan Lal Chadha and Sylvia Green 157 2.7 Conclusions and Recommendations Lisbeth Riis, Thomas Njuguna, Rebecca Raini, Berhard Löhr and Mohamed Ali Bob 163

4 iv Contents

Page SECTION THREE Whiteflies as Vectors of Viruses in Legume and Vegetable Mixed Cropping Systems in the Tropical Lowlands of Central America, Mexico and the Caribbean Chapter 3.1 Introduction Francisco Morales 173 3.2 Mexico Francisco Morales, Rafael Rivera-Bustamante, Rafael Salinas Pérez, Irineo Torres-Pacheco, Raúl Díaz Plaza, Wilson Avilés Baeza and Genovevo Ramírez Jaramillo 177 3.3 Guatemala Francisco Morales, Abelardo Viana, Margarita Palmieri, Mónica Orozco and René Ruano 188 3.4 El Salvador Francisco Morales, Carlos Pérez, Priscila Henríquez, Reina de Serrano, Leopoldo Serrano and Claudio Nunes 197 3.5 Honduras Francisco Morales, María Mercedes Doyle, Rogerio Trabanino, Carolina Nolasco, Elsa Barrientos, Luís Jara and Norman Escoto 204 3.6 Nicaragua Francisco Morales, Alberto Sediles, Aurelio Llano and Falguni Guharay 210 3.7 Costa Rica Francisco Morales, Luko Hilje, Juan Vallejos, Guillermo Sibaja, Carlos Araya and Rodolfo Araya 217 3.8 Panama Francisco Morales, Orencio Fernández and José Guerra 222 3.9 Haiti Francisco Morales, Jackson Donis and Emmanuel Prophete 226 3.10 Cuba Francisco Morales, Gloria González, Carlos Murguido, Ana Echemendía, Yolanda Martínez, Yenín Hernández, Benito Faure and María Chailloux 230 3.11 Dominican Republic Francisco Morales, Emigdio Gómez, Augusto Villar and Julio Nin 237 3.12 Reproductive Crop Hosts of Bemisia tabaci (Gennadius) in Latin America and the Caribbean Pamela K. Anderson, Avas Hamon, Pilar Hernández and Jon Martin 243

v 5 Whiteflies and Whitefly-borne Viruses in the Tropics

Page Chapter 3.13 Using Molecular Techniques to Analyse Whitefly Species and Biotypes in Latin America Lee Calvert, Natalia Villarreal and Donald Frohlich 251 3.14 Conclusions and Recommendations Francisco Morales 263

SECTION FOUR Whiteflies as Pests of Annual Crops in the Tropical Highlands of Latin America 4.1 Introduction César Cardona 269 4.2 Colombia and Ecuador César Cardona, Aristóbulo López-Avila and Oswaldo Valarezo 274 4.3 Insecticide Resistance in Colombia and Ecuador César Cardona, Francisco Rendón, Isaura Rodríguez and Aristóbulo López-Avila 285 4.4 Conclusions and Recommendations César Cardona, Aristóbulo López-Avila and Oswaldo Valarezo 295

SECTION FIVE Special Topics on Pest and Disease Management 5.1 Sustainable Integrated Management of Whiteflies through Host Plant Resistance Anthony Bellotti, Joseph Tohme, Michael Dunbier and Gail Timmerman 303 5.2 Biological Control of Whiteflies by Indigenous Natural Enemies for Major Food Crops in the Neotropics Anthony Bellotti, Jorge Peña, Bernardo Arias, José María Guerrero, Harold Trujillo, Claudia Holguín and Adriana Ortega 313 5.3 Progress of Cassava Mosaic Disease in Ugandan Cassava Varieties and in Varietal Mixtures William Sserubombwe, Michael Thresh, James Legg and William Otim-Nape 324 5.4 Management of the Cassava Mosaic Disease Pandemic in East Africa James Legg, James Whyte, Regina Kapinga and James Teri 332 Conclusions Francisco Morales 339 Acronyms and Abbreviations 345

6 vi CIAT Publication No. 341

Integrated Pest and Disease Management Project and Communications Unit

Editorial assistance: Gladys Rodríguez

Production: Graphic Arts Unit, CIAT Oscar Idárraga (layout) Julio C. Martínez (cover design)

Printing: Arte Libro Impresores, Cali, Colombia Contents

Page Foreword vii Introduction Pamela K. Anderson 1

iii Whiteflies and Whitefly-borne Viruses in the Tropics

4 iv Contents

v 5 Whiteflies and Whitefly-borne Viruses in the Tropics

6 vi Foreword

Pamela Anderson*

This book is the product of the the main tropical and sub-tropical collective knowledge accumulated over regions of the world, has been a major 3 decades by hundreds of agricultural challenge to agricultural scientists in professionals and scientists involved in both developing and industrialized the fight against two of the worst pests nations. Obviously, their impact has of all times: whiteflies and whitefly- been far greater in the rural areas of transmitted viruses. Millions of farmers developing countries, where resource- affected by these pests have also poor farmers have lost both traditional contributed a great deal of valuable food crops (e.g., common bean and information regarding the nature of the cassava) and cash crops (e.g., tomato problem and the extent of the damage and peppers) to these pests. Most of the caused by these pests in important national agricultural research institutes food and industrial crops. Examples of and ministries of agriculture in the such are: cassava (Manihot esculenta developing countries affected by these Crantz), common bean (Phaseolus pests have implemented different vulgaris L.), sweetpotato (Ipomoea whitefly control measures, including batatas [L.] Lam.), lima bean legal measures, which constitute a (Phaseolus lunatus L. var. lunatus), unique event in the history of most mung bean (Vigna radiata [L.] developing countries where agriculture R. Wilczek), tomato (Lycopersicon in general is not regulated by the state. esculentum Mill.), peppers (Capsicum spp. L.), squash (Cucurbita moschata The research activities described in Duchesne), melon (Cucumis melo L.), this book provide a general view of the other cucurbits, cotton (Gossypium problem for agricultural scientists and hirsutum L.), tobacco (Nicotiana laypeople alike, and contributes a great tabacum L.), soybean (Glycine max [L.] deal of pertinent information on the Merr.), lettuce (Lactuca sativa L. var. nature and management of the problem capitata L.) and eggplant (Solanum in the different crops affected by melongena L.). whiteflies and whitefly-transmitted viruses. In doing so, the Tropical The devastating capacity of these Whitefly Integrated Pest Management pests, and their global distribution in Project (TWF-IPM) has paid considerable attention to agricultural sustainability, ecosystem health and human health issues, related to the appalling misuse * Former coordinator for Phase 1 of the of highly toxic pesticides associated with Tropical Whitefly IPM Project; currently Research Director at the Centro whitefly control practices in developing Internacional de la Papa (CIP), Lima, Peru. countries.

vii Whiteflies and Whitefly-borne Viruses in the Tropics

The information contained in the systems. Moreover, this knowledge following chapters will form the basis of should ultimately contribute to improve a worldwide effort aimed at the livelihoods of resource-poor farmers disseminating the knowledge affected by these pests throughout the accumulated on the nature of the tropics, by safeguarding their food whitefly/virus problems and the most staples and high-value crops, reducing efficient integrated pest and disease production costs and pesticide management measures available to contamination, and increasing their develop sustainable mixed cropping family income.

8 viii Introduction

Introduction

Pamela K. Anderson*

The Problem of Whiteflies such as potato (Solanum tuberosum L.), and Whitefly-Transmitted tomato (Lycopersicon esculentum Mill.) Viruses in the Tropics and common bean (Phaseolus vulgaris L.). In addition, B. tabaci is a vector of Although Mound and Halsey (1978) plant begomoviruses (Geminiviridae: catalogued 1156 species of whiteflies Begomovirus). These whitefly- (Homoptera: Aleyrodidae), only a transmitted viruses (WTVs) are among limited number of whitefly species are the most destructive plant viruses; considered pests of economic early virus infection often results in importance. Two whitefly species are total crop loss. key pests throughout the tropics: Bemisia tabaci (Gennadius) and Although the problems caused by Trialeurodes vaporariorum (Westwood). B. tabaci, both as pest and vector, have Five additional whitefly species are been recognized for more than considered as important pests in 100 years, serious damage had been specific regions: Aleurotrachelus socialis limited to a handful of crops in (Bondar), Trialeurodes variabilis particular geographic areas. This (Quaintance), Bemisia tuberculata scenario has changed over the past (Bondar), Trialeurodes abutiloneus 2 decades. The known WTVs have (Haldman) and Aleurocanthus woglumi extended their geographic range and (Ashby). other WTVs are emerging in new crops and geographic zones, globally. Whiteflies are phloem (sap) feeders. Whitefly infestations now have become They cause direct damage in some severe in both traditional and non- hosts by extracting large quantities of traditional food and industrial crops sap. The honeydew that they excrete, throughout the tropics. as a result of the copious sap intake, serves as substrate for sooty mould In Africa, cassava mosaic disease fungi, which can also damage hosts by (CMD) was first described in 1894 from blocking photosynthesis. Sooty mould Tanzania (Warburg, 1894). In East can discolour harvestable fruits and Africa, the disease was not reported to fibre, affecting the quality of produce, cause serious losses until the 1920s. In and can cause plant death in crops West Africa, it was first recorded in the coastal areas of Nigeria, Sierra Leone and Ghana in 1929 and had spread * Former coordinator for Phase 1 of the northward by 1945. By 1987, CMD had Tropical Whitefly IPM Project; currently Research Director at the Centro been reported from all countries in Internacional de la Papa (CIP), Lima, Peru. Africa producing cassava (Manihot

1 Whiteflies and Whitefly-borne Viruses in the Tropics esculenta Crantz) (Fauquet and Central America and the Caribbean Fargette, 1990). The disease had been (Morales, 1994). Yield loss from either caused by African cassava mosaic virus BGMV or BGYMV is often 100% (ACMV) in South Africa and Africa west because of the high incidence of flower of the Rift Valley, and by East African abortion and the malformation of pods cassava mosaic virus (EACMV) in in infected plants (Morales and Madagascar and Africa east of the Rift Niessen, 1988). BGMV and BGYMV are Valley (Hong et al., 1993; Swanson and considered to be the limiting biotic Harrison, 1994). ACMV and EACMV are constraint to bean production in Latin both B. tabaci-transmitted America (Gálvez and Morales, 1989). begomoviruses. Then, in the late 1980s, tomato- In 1988, an extremely severe form producing areas in Latin America and of CMD was reported from north- across the tropics began to suffer from eastern Uganda. Infected cassava high incidences of whitefly-borne plants showed severe disease symptoms begomoviruses with devastating and produced little or no yield. The economic consequences. While no epidemic caused total crop failure in formal assessment studies of crop loss north-eastern Uganda (Otim-Nape et had been undertaken for the tomato al., 1997). Each year since 1988, the diseases caused by begomoviruses, the epidemic moved progressively empirical data were impressive. One of southwards along a broad front at a the more complete data sets came from rate of approximately 20 km per year the Dominican Republic (Polston and (Otim-Nape et al., 2000). By 1995, the Anderson, 1997). In 1988, multiple epidemic had reached Kenya (Otim- begomoviruses began to affect tomato Nape et al., 2000), with losses production in the Dominican Republic. estimated at 140,000 tons of cassava Crop damage from 1988 to 1995 per year, conservatively worth ranged from 5% to 95%. Economic US$14 million (Legg, 1999). The severe losses in 1988 were estimated at outbreak in Uganda and neighbouring US$10 million dollars, with losses from countries has been associated with a 1989 to 1995 totalling an estimated novel recombinant (EACMV+ACMV) US$50 million dollars. begomovirus, currently recognized as EACMV-Ug (Pita et al., 2001). These scenarios were being repeated across the tropics, with Soon after, Latin America was significant consequences to food struggling with Bemisia-transmitted security and poverty alleviation efforts bean and tomato viruses. Bean golden as well as human and ecosystem mosaic virus (BGMV) was first described health. in Brazil in the early 1960s as a minor disease (5%-10% incidence) of common Food security bean in the state of São Paulo (Costa, In Africa, cassava is one of the most 1965). Since then, BGMV has become a widely grown staple crops and second widespread problem in Brazil and now only to maize (Zea mays L.) in terms of is known to cause epidemics in north- calorie intake in sub-Saharan Africa. western Argentina and eastern Bolivia. About 200 million people, or 40% of Simultaneously, another begomovirus, the sub-Saharan Africa population, rely Bean golden yellow mosaic virus on cassava for their well-being. In some (BGYMV), which causes similar countries, people derive approximately symptoms, was ravaging common bean 1000 calories a day, or 50% of daily production in over 12 countries of food intake, from cassava (IITA, 1988).

2 Introduction

In addition to the yield losses diseases. Morales (1992) estimated that caused by CMD, an estimated by the early 1990s, approximately 150,000 ha of cassava-growing land 2,500,000 ha were under attack by was abandoned in Uganda during the BGMV/BGYMV and that at least height of the epidemic; that is, 1 million additional hectares could not equivalent to over 2.2 million metric be planted to beans each year because tons (US$440 million). This caused of the possibility of total yield losses, food shortages and famine in a number mainly during the dry seasons, when of districts, particularly in the eastern whitefly populations peak. Throughout and northern regions where the crop Central America and the Caribbean was the principle staple food (Otim- Basin, figures for crop damage Nape, 1997). indicated that the BGYMV infection was devastating. And there was continuing In Latin America, common bean is consensus that BGYMV was the major one of the main staple foods, biotic factor limiting bean production in particularly among the rural and urban Latin America (Morales, 1994). poor. In Central America, beans are the most important source of protein, Poverty alleviation usually being consumed three times a Many resource-poor producers in the day. Central America, despite its small tropics modified their traditional area (498,368 km2), devotes twice as cropping systems in the 1980s, to much of its geographical area to the incorporate non-traditional cash and cultivation of beans (735,00 ha), when export crops. The shift from compared to major bean producers subsistence to commercial agriculture such as Brazil (>5,000,000 ha). Beans is an opportunity to alleviate poverty in also are produced in some Caribbean rural areas. Smallholder producers islands, such as Cuba (26,000 t), the across the tropics have begun to Dominican Republic (55,000 t) and cultivate cash crops such as tomato Haiti (56,000 t) where they play an (Lycopersicon esculentum), pepper important nutritional role in the diet of (Capsicum spp. L.), melon (Cucumis the lower socio-economic classes. melo L.), watermelon (Citrullus lanatus Smallholdings cultivated by farmers [Thunb.] Matsum. & Nakai), squash with limited resources are (Cucurbita moschata Duchesne), grape characteristic of bean production in (Vitis vinifera L.), okra (Abelmoschus Central America and the Caribbean. In esculentus [L.] Moench), cucumber El Salvador, for instance, 85% of the (Cucumis sativus L. var. sativus), bean producers cultivate less than cabbage (Brassica oleracea L.), eggplant 14 ha, and 50% of these producers (Solanum melongena L.), broccoli have holdings of less than 3.5 ha. (Brassica oleracea L. var. italica Plenck) and ornamentals. However, most of Despite the large area planted to these horticultural crops are beans in Central America, average particularly susceptible to pests productivity is low (495 kg/ha) previously unknown to most small- compared to the average yield expected scale farmers. (over 1500 kg/ha) in most bean- producing regions of the USA and other In the absence of adequate temperate countries in the world. The technical assistance for small-scale main factor identified as responsible for farmers in the tropics, the new pest the low bean productivity in Latin problems were primarily dealt with by a America has been the incidence of myriad of pesticides applied biotic constraints, particularly indiscriminately. To further complicate

3 Whiteflies and Whitefly-borne Viruses in the Tropics this situation, a new biotype (B) of Human and ecosystem health B tabaci . was introduced in the The reliance of farmers on Americas in the early 1990s. This agrochemicals to protect their cash biotype proved to be far more crops against whitefly pests and polyphagous and fecund that the whitefly-borne viruses has resulted in original A biotype and has been an the systematic destruction of natural important factor in most of the major enemies that were once effective in outbreaks of whiteflies and whitefly- providing natural control, whitefly borne viruses. During the past decade, populations with high levels of the outbreak of whitefly pests and the insecticide resistance and the creation epidemics caused by whitefly- of new secondary pest problems. The transmitted begomoviruses in cash rejection of contaminated produce in crops has been often devastating. international markets has led to the Pesticide abuse is the norm in all sale of highly contaminated food whitefly-stricken regions of the tropics, products in developing countries, with which increases production costs and obvious detrimental health disqualifies contaminated (pesticide consequences for rural and urban residues) produce for export. consumers alike. Whiteflies provide a classic example of the pesticide At present, many of the small-scale treadmill. Insecticide abuse has farmers that attempted to diversify become a serious threat to the their cropping systems have failed to environment as well as a health hazard do so because of whitefly-associated to producers and consumers. problems. Prime agricultural land remains unexploited in many About 10% of the world’s developing countries during the dry population lives in the main highlands seasons, when whitefly populations and mountainous areas of the peak, despite the availability of water developing world (the Andes of South (irrigation districts) in these areas. The America, the Africa ranges and the welfare of developing countries and Himalayas). The management of their low-income citizens is tightly resources in those ecosystems affects linked to the existence of cash-earning an additional 40% of the world’s commodities, principally agricultural population, which inhabits adjacent products. areas such as Inter-Andean valleys in the Andean Zone. Traditionally, people The main force driving whitefly in the highlands have been outbreaks and crop losses due to marginalized from major development whitefly attack and WTVs is the lack of efforts, with significant repercussions qualified technical assistance to small- on poverty, migration and social unrest and medium-scale farmers in the as well as environmental deterioration tropics. The Tropical Whitefly in situ and downstream (IFPRI, 1995). Integrated Pest Management (TWF-IPM) Project is currently entering a final phase when most of the information One major issue regarding the presented in this book and the lessons welfare of those living in highland areas learned during the validation of IPM of Latin America is environmental practices in Phase II will be translated degradation caused by excessive into practical recommendations for pesticide use. It is well known that resource-poor farmers throughout the pesticide consumption in the tropics. developing world is increasing rapidly. It was thought initially that increases

4 Introduction in pesticide use in Latin America were ingredient per hectare per season), a mainly due to the growth of plantation crop that has been known traditionally crops, an important source of export as the worst offender in terms of revenue (Bellotti et al., 1990). However, insecticide use. And, of the insecticides as pointed out by Whitaker (1993), the utilized for whiteflies, 78% were developing country share of the world classified in toxicological category I agrochemical usage, currently valued (highly toxic) and most often applied in at US$10.6 billion, was forecast to rise mixture with other insecticides, broad- from 19% in 1988 to an estimated 35% spectrum fungicides and foliar by the year 2000. Much of this fertilizers. Farmers usually did not take projected growth was attributed to a precautions when handling pesticides; wider and more intensive use of up to 24% of those surveyed admit that chemical protection by smallholder they have been intoxicated at least farmers. Unfortunately, highly toxic once in the prior 10 years (CIAT, 1994). insecticides represent up to 45% of pesticide use in the developing world. In the east African highlands, a contributing factor to environmental In the early 1980s, the greenhouse degradation was the indiscriminate whitefly, Trialeurodes vaporariorum and/or excessive use of broad- (Westwood), became a very serious pest spectrum insecticides as part of the of several hillside-grown crops in the intensive crop protection system in Andes. Major outbreaks occurred in industrial crops, especially cotton, and 1987, 1991 and 1994 in selected areas to some extent on plantation crops of Colombia, northern Ecuador and the such as coffee (Coffea spp. L.) and tea Constanza Valley in Dominican (Camellia sinensis [L.] Kuntze). The Republic. For example, insecticide use problems of pesticide resistance and by small-scale bean farmers, which pest resurgence although initially was negligible until 1975-77 created on these target crops had (Schoonhoven and Cardona, 1980), has extended to the other seasonal crops increased steadily and became grown in rotation and/or combination excessive in Colombia and Ecuador. with the plantation crops, as in the Surveys conducted in two regions of case of whiteflies. Colombia and the northern part of Ecuador (CIAT, 1994) revealed that The increasing importance of 100% of 893 farmers surveyed sprayed horticultural production also brought their crops in an attempt to control the with it the attendant strategy of greenhouse whitefly. Highest pesticide-based protection in an effort insecticide use occurred in the to harvest damage-free produce. Sumapaz region of Colombia where Intensive use of pesticides in vegetable farmers make an average of crops had become quite common in 11.1 applications per season. large areas of Sudan, Uganda, Kenya, Tanzania and Zimbabwe. In a survey of Most alarming was the fact that vegetable farmers in Kenya it was some growers sprayed their crops up to found that the majority perceived that 24 times in a crop cycle of they would lose up to 90% of their 90-100 days, that is, every 3 to 4 days. harvest if they did not use pesticides. Insecticide abuse on beans in Colombia By the early 1990s, these vegetable (5.6 kg active ingredient per ha per farmers were applying up to 19 sprays season) could be compared to a season, with a significant proportion insecticide consumption for cotton of farmers spraying 9-12 times per (Gossypium hirsutum L.) (6.2 kg active season (KARI-GTZ, 1994). It was

5 Whiteflies and Whitefly-borne Viruses in the Tropics critical to begin work in eastern Africa initiatives included the creation of the in hopes of intervening with alternative SP-IPM. Whitefly and WTV research had whitefly crop protection measures, been ongoing at several of the IARCs before the whitefly pests and WTVs since the 1970s. Due to the growing reach the severity witnessed in the importance of the whitefly and WTV Neotropics. problem, the TWF-IPM Project was the first inter-centre project approved There was an urgent need to within the new programme. The develop IPM systems that would reduce International Center for Tropical pesticide use and help re-establish the Agriculture (CIAT, the Spanish ecological equilibrium by means of acronym) was designated as the non-chemical approaches to whitefly convening centre to organize the Inter- management. For whitefly pests this Centre Whitefly IPM Task Force and to implies identifying the principal crop formulate the Inter-Centre proposal on hosts, establishing economic injury Sustainable Integrated Management of levels (EILs) and developing new Whiteflies as Pests and Vectors of Plant approaches to maintain whiteflies Viruses in the Tropics (the CGIAR below the EIL. For whitefly vectors, TWF-IPM Project). however, the traditional IPM approach will not suffice. Vectors must be A Task Force Meeting was held at studied and managed within an CIAT in Cali, Colombia, from epidemiological framework, that is, 13-15 February 1996. The objectives of study and analysis of the WTV system the meeting were to discuss: (a) the with IPM intervention strategies goal, purpose, outputs and activities resulting from the epidemiological that should be proposed for the project; analysis. and (b) a structure for the global Whitefly IPM project, as well as how to link and coordinate the institutions that Origin and Organization would be involved in the project. of the CGIAR TWF-IPM Project The Task Force Meeting included 24 participants representing IARCs, National Agricultural Research Systems The Systemwide Programme on (NARS) and Advanced Research Integrated Pest Management Institutions (ARIs). After considerable (SP-IPM) discussion on the nature of the whitefly The Consultative Group for problem, the Task Force agreed that it International Agricultural Research was possible to define three whitefly (CGIAR) consists of 15 International problems that should be prioritized: Agricultural Research Centres (IARCs) located around the world. Historically, (1) Whiteflies as vectors in mixed each Centre has worked in a relatively cropping systems in low to mid independent and autonomous fashion. altitudes of the tropics; During the 1990s, the CGIAR (2) Whiteflies as pests in mixed Secretariat recognized that significant cropping systems in tropical benefits could be derived by pooling highlands; and human capital, infrastructure and (3) Whiteflies as vectors and pests of economic resources to jointly tackle the semi-perennial cassava. research problems that numerous centres were working on The first problem focused on independently. New inter-centre whiteflies as vectors of plant viruses in

6 Introduction annual crops, especially legumes and feeding damage and WTVs. The Task vegetables, in the tropical lowlands. Force agreed to organize the project The second problem focused on around six outputs: whiteflies as direct pests in annual crops in the highlands. And the third (1) Formation of an international problem focused on whitefly pests and network for research on whiteflies vectors in a semi-perennial crop, and WTVs in the tropics; cassava. (2) Characterization of whitefly problems in order to prioritize Based on the initial Whitefly IPM critical target areas; Task Force Meeting, and within the CG (3) Improvement of the understanding framework for an eco-regional problem of whitefly pest and disease approach (Bouma et al., 1995), the dynamics in critical target areas; project was structured into six sub- (4) Development and testing of IPM projects: strategies and tactics; (5) Strengthening of NARS research (1) Bemisia tabaci as a virus vector in capacity, policy formulation and cassava and sweet potato in sub- IPM implementation; and Saharan Africa – led by IITA; (6) Assessment of project impact. (2) Bemisia tabaci as a virus vector in mixed cropping systems of the Phase 1 of the CGIAR TWF-IPM Caribbean, Mexico and Central Project America – led by CIAT; The project was conceptualized in three (3) Bemisia tabaci as a virus vector in phases, to be carried out over 10 years. mixed cropping systems of Eastern Phase 1 of the project would focus and Southern Africa – led by the primarily on Output 1 (network International Centre of Insect formation) and Output 2 (problem Physiology and Ecology (ICIPE); characterization) as the basis for (4) Bemisia tabaci as a virus vector in Phase 2 work, with limited activities in mixed cropping systems of S.E. Output 3 (pest and disease dynamics) Asia – led by the Asian Vegetable and Output 4 (development and testing Research and Development Center of IPM strategies and tactics). The (AVRDC); Project was officially launched in 1997 (5) Trialeurodes vaporariorum as a and evolved into a pan-tropical direct pest in the tropical partnership. Five IARCs were included: highlands of Latin America – led by CIAT; and CIAT Centro Internacional de (6) Whiteflies as direct pests on Agricultura Tropical cassava in South America – led by CIAT. IITA International Institute of Tropical Agriculture There was consensus by the Task Force on the project goal, project AVRDC Asian Vegetable Research and purpose and project outputs for the Development Center TWF-IPM project. The project goal is to improve living conditions of rural ICIPE International Centre of Insect families through the effective Physiology and Ecology management of whiteflies, resulting in increased crop production and a safer CIP Centro Internacional de la environment. The project purpose is to Papa reduce crop losses due to whitefly

7 Whiteflies and Whitefly-borne Viruses in the Tropics

Twelve ARIs were also included: Bangladesh, Thailand, Malaysia, Vietnam, Indonesia and the NRI Natural Resources Institute Philippines. (UK) This collaborative partnership was JIC John Innes Centre (UK) possible due to the support of six donor partners: BBA Biologische Bundesanstalt für Land-und Fortwirtschaft DANIDA Danish International (Germany) Development Agency

CATIE Centro Agronómico Tropical de ACIAR Australian Centre for Investigación y Enseñanza International Agricultural (Costa Rica) Research

UA University of Arizona-Tucson USAID United States Agency for (USA) International Development

UFL University of Florida- MFAT Ministry of Foreign Affairs Gainesville (USA) and Trade (New Zealand, now New Zealand Aid) UW University of Wisconsin- Madison (USA) USDA- U.S. Department of (ARS) Agriculture-Agricultural MSU Montana State University Research Service (USA) DFID Department for International FSAC Florida State Arthropod Development (UK) Collection The CGIAR TWF-IPM Project was USDA- U.S. Department of launched in 1997 with the DANIDA- (ARS) Agriculture-Agricultural funded Phase 1 project on Sustainable Research Service Integrated Management of Whiteflies as CSIRO Commonwealth Scientific and Pests and Vectors of Plant Viruses in Industrial Research the tropics. This project covered Organisation (Australia) activities in Output 1 (network formation) and Output 2 C&F New Zealand Crop and Food (characterization) activities in Research 23 African and Latin American partner countries in Sub-projects 1, 2, 3 and And 31 National Agricultural 5. ACIAR then approved funding for the Research and Extension Systems project Sustainable Integrated (NARES) in Latin America, the Management of Whiteflies as Pests and Caribbean, Africa and Asia were Vectors of Plant Viruses in S.E. Asia, included from: which covered Output 2 (characterization) activities in the eight Mexico, Guatemala, Honduras, Asian countries in Sub-project 4. The El Salvador, Nicaragua, Costa Rica, USAID-funded project on Biological Panama, El Salvador, Cuba, Haiti, Control of Whiteflies by Indigenous Dominican Republic, Colombia, Natural Enemies for Major Food Crops Ecuador, Sudan, Benin, Ghana, Nigeria, in the Neotropics and the MFAT-funded Cameroon, Kenya, Uganda, Tanzania, project on Sustainable Integrated Malawi, Madagascar, Nepal, Sri Lanka, Management of Whiteflies through Host

8 Introduction

Plant Resistance contributed to Output research agenda for improved 2 (characterization) and Output 4 (IPM) understanding of pest and disease activities in Sub-project 6. USDA-ARS dynamics, IPM development and IPM signed a Scientific Cooperative implementation. Several of the project Agreement to link the ARS national coordinators have published review research on whitefly IPM research with articles (Polston and Anderson, 1997; the international effort. This agreement Legg, 1999; Morales, 2000; Bellotti and on Integrated Pest Management Arias, 2001; Morales, 2001; Morales Strategies for Whitefly-transmitted and Anderson, 2001; Morales and Viruses supported Output 1 (network Jones, 2004). Baseline data were formation) and preliminary Output 3 generated through extensive survey (disease dynamics) research activities work in each of the participating on epidemiology of whitefly-transmitted tropical countries. Surveys included begomoviruses. the collection and identification of biological specimens (whitefly species, The CMD epidemic in eastern whitefly biotypes, geminiviruses, and Africa was of particular concern to the B. tabaci natural enemies) and TWF-IPM Project for humanitarian collection and analysis of socio- reasons. The USAID Office of Foreign economic data and IPM practices, Disaster Assistance (OFDA) granted based on producer interviews. All special funding for the Emergency partners have presented preliminary Programme to Combat the Cassava results of the extensive survey work Mosaic Disease Pandemic in East and IPM measures in international, Africa. The objective of this disaster regional and national meetings. This assistance is to boost production of book is the first presentation of cassava in Uganda, Kenya and detailed results generated by the Tanzania and enhance both short- and Phase 1 projects contributing to the longer-term food security through the CGIAR TWF-IPM Project. For further implementation of an emergency information on specific methodologies program to multiply and disseminate used to produce the results published CMD-resistant cassava (Output 4 in this book, please contact the activities). And DFID funded the project information and communication on Promotion of and Technical Support assistant of the project for Methods of Controlling Whitefly- (www.tropicalwhiteflyipmproject.cgiar.org). borne Viruses in Sweet Potato in East Africa; sweetpotato (Ipomoea batatas References [L.] Lam.) served as the principal food security crop in the midst of the CMD Bellotti, A. C.; Arias, B. 2001. Host plant epidemic. resistance to whiteflies with emphasis on cassava as a case study. Crop Prot. 20:813-823. Whiteflies and Whitefly-Borne Viruses in the Tropics: Building a Bellotti, A. C.; Cardona, C.; Lapointe, S. Knowledge Base for Global Action L. 1990. Trends in pesticide use in reports the results from Phase 1 of the Colombia and Brazil. J. Agric. CGIAR TWF-IPM Project. The purpose Entomol. 7:191-201. of Output 2 (problem characterization) Bouma, J.; Kuyvenhoven, A.; Bouman, B. was to gather, review, generate and A. M.; Luyten, J. C. (eds.). 1995. analyse baseline data relevant to the Eco-regional approaches for diagnosis and characterization of sustainable land use and food whitefly and WTV problems in the production. Kluwer Academic tropics, in order to propose a sound Publishers, Dordrecht, NL.

9 Whiteflies and Whitefly-borne Viruses in the Tropics

CIAT (Centro Internacional de Agricultura Morales, F. J. 1992. Annual progress Tropical). 1994. Annual report. report (1992) and five-year report CIAT-Bean Program, Cali, CO. (1988-1992). International Center p. 171-179. for Tropical Agriculture (CIAT), Cali, CO. Costa, A. S. 1965. Three whitefly- transmitted virus diseases of beans Morales, F. J. (ed.). 1994. Bean golden in São Paulo, Brazil. FAO Plant mosaic: Research advances. Centro Prot. Bull. 13:121-130. Internacional de Agricultura Tropical (CIAT), Cali, CO. 193 p. Fauquet, C.; Fargette, D. 1990. African cassava mosaic virus: Etiology, Morales, F. J. (ed.). 2000. Bean golden epidemiology and control. Plant Dis. mosaic virus and other diseases of 74:404-411. common bean caused by whitefly- transmitted geminiviruses in Latin Gálvez, G.; Morales, F. J. 1989. Whitefly- America. Centro Internacional de transmitted viruses. In: Schwartz, Agricultura Tropical (CIAT), Cali, H. F.; Pastor-Corrales, M. A. (eds.). CO. 169 p. Bean production problems in the tropics. Centro Internacional de Morales, F. J. 2001. Conventional Agricultura Tropical (CIAT), Cali, breeding for resistance to Bemisia- CO. p. 379-406. transmitted geminiviruses. Crop Prot. 20:825-834. Hong, Y. G.; Robinson, D. J.; Harrison, B. D. 1993. Nucleotide sequence Morales, F. J.; Anderson, P. K. 2001. The evidence for the occurrence of three emergence and dissemination of distinct whitefly-transmitted whitefly-transmitted geminiviruses geminiviruses in cassava. J. Gen. in Latin America. Arch. Virol. Virol. 74:2437-2443. 146(3):415-441.

IFPRI (International Food Policy Research Morales, F. J.; Jones, P. G. 2004. The Institute). 1995. A 2020 vision for ecology and epidemiology of food, agriculture, and the whitefly-transmitted viruses in environment. International Latin America. Virus Res. 100: Conference, 13-15 June 1995, 57-65. Washington, DC. 145 p. Morales, F. J.; Niessen, A. I. 1988. IITA (International Institute of Tropical Comparative responses of selected Agriculture). 1988. Strategic plan Phaseolus vulgaris germplasm 1989-2000. Ibadan, NG. inoculated artificially and naturally with Bean golden mosaic virus. Plant KARI-GTZ (Kenya Agricultural Research Dis. 72:1020-1023. Institute - German Agency for Technical Cooperation). 1994. Crop Mound, L. A.; Halsey, S. H. 1978. protection measures of Kenyan Whitefly of the world: A systematic vegetable farmers and their use of catalogue of the Aleyrodidae pesticides, knowledge, attitude and (Homoptera) with host plant and practice survey. GTZ-Integrated natural enemy data. John Wiley, Pest Management Horticulture Chichester, GB. 340 p. Project Report, Nairobi, KE. 45 p. Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; Legg, J. 1999. Emergence, spread and Baguma, Y.; Ogwal, S.; Semakula, strategies for controlling the G. N.; Acola, G.; Byabakama, B.; pandemic of cassava mosaic virus Martin, A. 1997. Cassava mosaic disease in east and central Africa. virus disease in Uganda: The Crop Prot. 18:627-637. current pandemic and approaches to control. Natural Resources Institute (NRI), Chatham, GB. 65 p.

10 Introduction

Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; Schoonhoven, A. van; Cardona, C. 1980. Baguma, Y.; Ogwal, S.; Ssemakula, Insects and other bean pests in G. N.; Acola, G.; Byabakama, B.; Latin America. In: Bean production Colvin, J.; Cooter, R. J.; Martin A. problems: Disease, insect, soil and 2000. The current pandemic of climatic constraints of Phaseolus cassava mosaic virus disease in vulgaris. Centro Internacional de East Africa and its control. CAB Agricultura Tropical (CIAT), Cali, International, Wallingford, GB. CO. p. 363-412. 100 p. Swanson, M. M.; Harrison, B. D. 1994. Pita, J. S.; Fondong, V. N.; Sangaré, A.; Properties, relationships and Otim-Nape, G. W.; Ogwal, S.; distribution of cassava mosaic Fauquet, C. M. 2001. geminiviruses. Trop. Sci. 34:15-25. Recombination, pseudorecombination and Warburg, O. 1894. Die Kulturpflazen synergism of geminiviruses are usambaras. Mitt. Dtsch. Schutz. determinant keys to the epidemic of 7:131. severe cassava mosaic disease in Uganda. J. Gen. Virol. 82:655-665. Whitaker, M. J. 1993. The challenge of pesticide education and training for Polston, J. P.; Anderson, P. K. 1997. The tropical smallholders. Int. J. Pest emergence of whitefly-transmitted Manage. 39(2):117-125. geminiviruses in tomato in the Western Hemisphere. Plant Dis. 81:1358-1369.

11 blanca 12 SECTION ONE

Whiteflies as Vectors of Plant Viruses in Cassava and Sweetpotato in Africa BLANCA 14 Introduction

CHAPTER 1.1 Introduction

James Legg*

More than 1200 species of whitefly the earliest reports relating to damage (Homoptera: Aleyrodidae) have been in cotton (Gossypium hirsutum L.) described worldwide, of which roughly (Kirkpatrick, 1931) and cassava one quarter occurs in Africa (Mound (Manihot esculenta Crantz) (Kufferath and Halsey, 1978). Of these diverse and Ghesquière, 1932; Storey and whitefly species, only a few are of Nichols, 1938). By the 1950s, significant economic importance. These B. tabaci was reported causing major include: Aleurocanthus spiniferus losses in cotton as a result of (Quaintance), Aleurocanthus woglumi population resurgence of this species Ashby, Aleurodicus dispersus Russell, when it developed resistance to Aleurothrixus floccosus (Maskell), pesticides applied against other pests Aleyrodes proletella (Linnaeus), Bemisia (Joyce, 1955). And, by the 1980s, tabaci (Gennadius), Dialeurodes citri B. tabaci had been associated with a (Ashmead), Parabemisia myricae major pandemic of cassava mosaic (Kuwana), Siphoninus phillyreae disease (CMD) in East Africa (Otim- (Haliday), Trialeurodes vaporariorum Nape et al., 1997; Legg, 1999). (Westwood) and Vasdavidius indicus (David and Subramaniam) Russell. Principal among these, however, is Cassava and Sweetpotato B. tabaci, which is a pest on a wide in Sub-Saharan Africa range of field crops throughout the continent. According to international production statistics (FAO, 2004), Africa B. tabaci was first described from produces almost half of the world’s tobacco (Nicotiana tabacum L.) in crop of cassava and East Africa is one Greece and is thought to have of the world’s most important areas originated from the Old World (Frohlich for growing sweetpotato (Ipomoea et al., 1996). It has been recognized as batatas [L.] Lam). Nigeria is the an important agricultural pest in Africa world’s leading producer of cassava, since at least the 1930s, with several of while Uganda ranks third, worldwide, in total sweetpotato production. These two crops are of major importance to sub-Saharan Africa * International Institute of Tropical and play vital roles in sustaining food Agriculture-Eastern and Southern Africa security, particularly in view of their Regional Center (IITA-ESARC), Kampala, ability to provide reasonable yields Uganda, and Natural Resources Institute, University of Greenwich, Chatham Maritime, even under conditions of poor soil Kent, UK. fertility or drought (Jennings, 1970).

15 Whiteflies and Whitefly-borne Viruses in the Tropics

Cassava was brought to Africa in transmissible (Zimmerman, 1906). It the sixteenth century by Portuguese was not until the 1970s, however, that seafarers and, perhaps because of this the viral etiology was confirmed (Bock relatively recent introduction, has few and Woods, 1983). CMD is caused by major pests and diseases. This is in begomoviruses (Bock and Woods, contrast to the wide diversity of pests 1983; Hong et al., 1993), which are and diseases that affect this crop in its transmitted by B. tabaci (Storey and native South America (Bellotti et al., Nichols, 1938; Chant, 1958; Dubern, 1994). The two most important 1979). Cassava mosaic begomoviruses arthropod pests of cassava in Africa, occurring in Africa are thought to be the cassava mealybug (Phenacoccus indigenous to that continent; it is manihoti Matile-Ferrero) and the assumed that they moved into cassava cassava green mite (Mononychellus from wild host plants following the tanajoa [Bondar]) were both introduced introduction of cassava from South from South America in the early 1970s America (Swanson and Harrison, (Yaninek and Herren, 1988; 1994). Neuenschwander, 1994). They spread rapidly to all cassava-producing zones Cassava was subsidiary in of mainland Africa during the 1980s importance to other staple crops until and 1990s but both have been the early part of the twentieth century controlled successfully through the but promotion of the crop as a famine implementation of continent-wide reserve by colonial authorities in the biological control programs (Herren and 1920s and 1930s rapidly led to it being Neuenschwander, 1991; Yaninek et al., more widely grown and consumed 1993). (Hillocks, 2002). It appears that this increase in cultivation area and The principal diseases of cassava intensity, coupled possibly with the in Africa are CMD and cassava unrestricted movement of germplasm bacterial blight. The latter, like the two between countries, catalyzed the arthropod pests described above, was continent-wide spread of CMD. New introduced into Africa in the 1970s and occurrences as well as the increasing spread rapidly to all cassava-growing importance of CMD were reported from areas (Boher and Verdier, 1994). The numerous countries in East and West disease is caused by the bacterium Africa, including Sierra Leone Xanthomonas axonopodis pv. manihotis (Deighton, 1926), Uganda (Hall, 1928), (Xam). Although, under certain Ghana (Dade, 1930), Nigeria (Golding, circumstances, damage may be very 1936) and Madagascar (François, severe and result in total crop loss, 1937). Losses due to CMD in Africa outbreaks with serious effects are have been estimated at between usually localized. Cassava bacterial 12 and 23 million tons annually, blight is therefore less economically equivalent to between US$1200 and important than CMD. US$2300 million (Thresh et al., 1997). Following the success of biological CMD was first described more than control in reducing losses caused by a century ago (Warburg, 1894) from cassava mealy bug and cassava green what is modern-day Tanzania. It was mite, CMD generally is regarded as the assumed from the early years of the most important biotic constraint on twentieth century that the disease was cassava production in Africa. caused by a virus, since no pathogens were visible on affected plants and the Sweetpotato in Africa is attacked condition was shown to be graft by a greater diversity of pests and

16 Introduction diseases than is cassava, although, as the pandemic has been associated with with cassava, few are of major a new, more virulent, cassava mosaic economic importance. The most begomovirus (Deng et al., 1997; Zhou important pests include the et al., 1997). Super-abundant B. tabaci sweetpotato weevil (Cylas spp.), the populations associated with the striped sweetpotato weevil (Alcidodes pandemic also have been shown to dentipes [Oliver]), the sweetpotato result from a synergistic interaction butterfly (Acraea acerata Hewitson), leaf with severely CMD-diseased cassava blight (Alternaria spp.) and sweetpotato plants, in which the diseased plants virus disease (SPVD). The latter has appear to promote B. tabaci population been ranked as the single most increase, possibly through the disease important constraint to sweetpotato improving the food quality of the plant production in the East African region for B. tabaci (Colvin et al., 1999). and is the most important disease However, evidence has been presented throughout Africa (Geddes, 1990). It for a link between a distinct B. tabaci results from co-infection of sweetpotato genotype cluster and the pandemic in plants with the aphid-borne Uganda (Legg et al., 2002). Sweetpotato feathery mottle virus (SPFMV) and the B. tabaci-borne In contrast to CMD, knowledge of Sweetpotato chlorotic stunt virus SPVD and the role of B. tabaci in the (SPCSV) (Schaefers and Terry, 1976; ecology of the disease is extremely Gibson et al., 1998). limited. Studies have been restricted largely to virus characterization, etiology and transmission, virus-virus Status of CMD and SPVD interactions, and yield loss Research assessments (Schaefers and Terry, 1976; Hahn, 1979; Gibson et al., 1997; Since its description more than a 1998). Although the disease is known century ago, CMD has received to occur widely in sub-Saharan Africa, significant research attention. Fauquet virtually nothing is known about its and Fargette (1990) and Thresh et al. prevalence in the major sweetpotato (1994; 1998) have reviewed a wide cultivation areas and at the inception range of studies on the biological and of this study published incidence data molecular characterization, etiology, were available only for Uganda (Aritua epidemiology, yield loss and control of et al., 1998). these viruses. Fishpool and Burban (1994) and Legg (1994) have reviewed Although significant progress had studies on B. tabaci as the vector of been made previously in characterizing cassava mosaic geminiviruses, whitefly-transmitted viruses and including work on biotype whiteflies associated with CMD and characterization, population dynamics, SPVD, at the time of launching the virus transmission and movement present study, significant gaps in within cassava fields. The emergence of understanding remained (Table 1). This a severe form of CMD in the 1990s was most apparent for B. tabaci, where (Otim-Nape et al., 1997; Legg and very little research other than one-time Ogwal, 1998; Legg, 1999) has had a assessments of abundance and limited devastating effect on cassava population dynamics studies had been cultivation across a large area of East done outside Uganda and the Ivory and Central Africa. Rapid expansion of Coast.

17 Whiteflies and Whitefly-borne Viruses in the Tropics B. tabaci Farmer Natural enemies ase (SPVD) research at time of yield lossesyield perceptions of CMD/SPVD CMD/SPVD epidemiology a B. tabaci Research topic abundance CMD/SPVD Bemisia tabaci characterisation prevalence Virus characterisation KenyaTanzaniaMalawiMadagascarGhana (+) (+)BeninNigeria (+) (+)Cameroon (+)Kenya (+) - -Tanzania (+) (+) -Malawi -Madagascar - (+) + (+) - + ------+ (+) - + - + + + - - - - + + (+) + - + + + - - - - - + - - - - + + ------+ + - - - - (+) - - (+) (+) (+) ------starting present study in Project target countries of sub-Saharan Africa. Disease Country CMD Uganda (+)SPVD (+) Uganda + + + - + + + - (+) (+) (+) + - - a. + Published literature available; (+) Limited study published; - No published available. Table 1. Scope of published work in sub-Saharan Africa for key fields cassava mosaic disease (CMD) and sweetpotato virus dise

18 Introduction

Sub-project on Whiteflies farmers, although security concerns as Vectors of Viruses of precluded the participation of some of Cassava and Sweetpotato the major producers of each crop. The in Sub-Saharan Africa nine African countries that were extensively surveyed—Ghana, Benin, Within the framework of the Tropical Nigeria, Cameroon, Uganda, Tanzania, Whitefly Integrated Pest Management Kenya, Malawi and Madagascar— (TWF-IPM) Project, the purpose of the together represented 62% of cassava diagnostic phase of the sub-project was and 56% of sweetpotato production in to gather, generate and analyse Africa (FAO, 1999). Zambia was baseline data relevant to the diagnosis involved in the project only to a limited and characterization of whitefly extent, implementing a sweetpotato problems and whitefly-transmitted virus diagnostic survey in collaboration virus problems in cassava and with the Natural Resources Institute, sweetpotato. The rationale behind the UK. Sweetpotato work was carried out diagnostic phase was therefore to only in eastern and southern Africa, provide essential baseline information because the crop is not grown widely on CMD, SPVD and whiteflies on in West Africa. Table 2 summarizes the cassava and sweetpotato, to serve as roles of each of the partners involved the basis for the development and in the diagnostic phase. implementation of targeted and appropriate IPM strategies in Implementation of the work subsequent phases of the project. plan, results and analysis Diagnostic surveys were conducted The sub-project involved according to protocols set out in the collaboration among three international standardized methodology agreed agricultural research centres, one among the project partners. In each national agricultural research country, three or four target areas were organization in each of nine African identified for the survey. The criteria countries, and three other research for selection of these areas were that institutions outside Africa (Table 2). they should be major root crop- The inclusion of nine countries at this producing zones and preferably should stage of the project and the use of a have contrasting agro-ecological standardized protocol provided a characteristics. In each ecozone, the unique opportunity to obtain data that sites were selected at about 20-km could be compared from country to intervals along roads. Where no country and region to region. The scope cassava farms were encountered at the of the study also facilitated the first 20-km point, the next suitable site systematic collection and thereafter was selected. Field data and characterization of whitefly-transmitted biological specimens were collected viruses, whitefly and whitefly natural from 3-6 month-old cassava plantings enemy specimens from cassava and on surveyed farms and producers were sweetpotato in sub-Saharan Africa. interviewed using the standardized project questionnaires. Collaboration and selection of target areas Between 10 and 42 locations were Countries were identified for used for sampling in each target area, participation in the sub-project and at each location a single set of primarily on the basis of the samples and data was collected. The importance of cassava and/or data set comprised: (1) a producer sweetpotato cultivation to their questionnaire relating to perceptions

19 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 2. Partners involved in the diagnostic phase of the sub-project on Whiteflies as Vectors of Viruses of Cassava and Sweetpotato in sub-Saharan Africa and their roles.

Institution Country Rolea

International agricultural research centres: International Institute of Tropical Agriculture (IITA) Uganda/Benin Co-ordination Centro Internacional de la Papa (CIP) Uganda SPVD epidemiology International Centre of Insect Physiology and Kenya Whitefly and natural Ecology (ICIPE) enemy species identification

National agricultural research organizations/institutes: Plant Protection and Regulatory Services Directorate Ghana Diagnostic survey (PPRSD) Institut National de Recherche Agronomique Benin Diagnostic survey du Bénin (INRAB) National Root Crops Research Institute (NRCRI) Nigeria Diagnostic survey Institut de Recherche Agronomique (IRA) Cameroon Diagnostic survey National Agricultural Research Organisation (NARO) Uganda Diagnostic survey Tanzania Root and Tuber Crops Programme (TRTCP) Tanzania Diagnostic survey Kenya Agricultural Research Institute (KARI) Kenya Diagnostic survey Chitedze Research Station (CRS) Malawi Diagnostic survey Centre National de Recherche Appliquée au Développement Rural (FOFIFA) Madagascar Diagnostic survey Mount Makulu Research Station Zambia Diagnostic survey (SPVD only)

Specialized research organizations: Natural Resources Institute (NRI) United Kingdom Epidemiology of CMD and SPVD Biologische Bundesanstalt für Land und Germany Sweetpotato virus Fortwirtschaft (BBA) diagnostics John Innes Centre (JIC) United Kingdom Molecular diagnostics of cassava viruses and whiteflies a. CMD, cassava mosaic disease; SPVD, sweetpotato virus disease. and management of whiteflies and section. Data from the producer whitefly-transmitted virus problems; questionnaires for all countries were (2) biological samples of whiteflies, entered into a single database from virus-diseased leaf and stem tissues, which they could be summarized and and whitefly natural enemies; and queried. Field data from all countries (3) field assessments of disease were summarized and integrated to incidence, virus symptoms and whitefly develop maps (Chapter 1.14, this abundance. Before carrying out the volume). Studies conducted in survey, lead scientists in each country partnership with research institutions were trained in survey methods. outside Africa were initiated in early Surveys were conducted between 1998 and completed by mid-1999; November 1997 and September 1998. results are reported later in this section The results are reported in the (Chapters 1.11, 1.12 and 1.13) or in subsequent country chapters in this Section 5 (Chapter 5.3).

20 Introduction

References Fauquet, C.; Fargette, D. 1990. African cassava mosaic virus: Etiology, Aritua, V.; Adipala, E.; Carey, E. E.; epidemiology and control. Plant Dis. Gibson, R. W. 1998. The incidence 74:404-411. of sweetpotato virus disease and virus resistance of sweetpotato Fishpool, L. –D .C.; Burban, C. 1994. grown in Uganda. Ann. Appl. Biol. Bemisia tabaci: The whitefly vector 132:399-411. of African cassava mosaic geminivirus. Trop. Sci. 34:55-72. Bellotti, A. C.; Braun, A. R.; Arias, B.; Castillo, J. A.; Guerrero, J. M. 1994. FAO (Food and Agriculture Organization Origin and management of of the United Nations). 1999. neotropical cassava arthropod Cassava production data 1998. pests. Afr. Crop Sci. J. 2:407-418. Available in: http://www.fao.org

Bock, K. R.; Woods, R. D. 1983. Etiology FAO (Food and Agriculture Organization of African cassava mosaic disease. of the United Nations). 2004. Plant Dis. 67:944-955. Cassava production data 2003. Available in: http://www.fao.org Boher, B.; Verdier, V. 1994. Cassava bacterial blight in Africa: The state François, E. 1937. Un grave peril: La of knowledge and implications for mosaïque du manioc. Agron. Colon. designing control strategies. Afr. 26:33-38. Crop Sci. J. 2:505-509. Frohlich, D. R.; Brown, J. K.; Bedford, I.; Chant, S. R. 1958. Studies on the Markham, P. 1996. Mitochondrial transmission of cassava mosaic 16S ribosomal subunit as a virus by Bemisia spp. (Aleyrodidae). molecular marker in Bemisia, and Ann. Appl. Biol. 46:210-215. implications for population variability. In: Gerling, D.; Mayer, Colvin, J.; Otim-Nape, G. W.; Holt, J.; R. T. (eds.). Bemisia 1995: Omongo, C.; Seal, S.; Stevenson, P.; Taxonomy, biology, damage, control Gibson, R.; Cooter, R. J.; Thresh, J. and management. Intercept, M. 1999. Factors driving the current Andover, GB. p. 143-145. epidemic of severe cassava mosaic disease in East Africa, In: Procs. Geddes, A. M. 1990. The relative VIIth International Plant Virus importance of crop pests in sub- Epidemiology Symposium, 11-16 Saharan Africa. Bulletin no. 36, April 1999, Almeria, ES. p. 76-77. Natural Resources Institute (NRI), Chatham, GB. Dade, H. A. 1930. Cassava mosaic. In: Yearbook. Department of Gibson, R. W.; Mwanga, R. O. M.; Kasule, Agriculture, GH. p. 245-247. S.; Mpembe, I.; Carey, E. E. 1997. Apparent absence of viruses in Deighton, F. C. 1926. Annual report of most symptomless field-grown the Lands and Forestry sweetpotato in Uganda. Ann. Appl. Department. Freetown, SL. p. 1-2. Biol. 130:481-490.

Deng, D.; Otim-Nape, G. W.; Sangare, A.; Gibson, R. W.; Mpembe, I.; Alicai, T.; Ogwal, S.; Beachy, R. N.; Fauquet, Carey, E. E.; Mwanga, R. O. M.; C. M. 1997. Presence of a new virus Seal, S. E.; Vetten, H. J. 1998. closely associated with cassava Symptoms, aetiology and serological mosaic outbreak in Uganda. Afr. J. analysis of sweetpotato virus Root Tuber Crops 2:23-28. disease in Uganda. Plant Pathol. 47:95-102. Dubern, J. 1979. Quelques propriétés de la mosaïque africaine du manioc. I: Golding, F. D. 1936. Cassava mosaic in La transmission. J. Phytopathol. Southern Nigeria. Bull. Dept. Agric. 96:25-39. 11:1-10. 21 Whiteflies and Whitefly-borne Viruses in the Tropics

Hahn, S. K. 1979. Effects of sweetpotato Legg, J. P.; Ogwal, S. 1998. Changes in virus disease (SPVD) on growth and the incidence of African cassava yield of sweetpotato. Exp. Agric. mosaic geminivirus and the 15:253-256. abundance of its whitefly vector along south-north transects in Hall, F. W. 1928. Annual report. Uganda. J. Appl. Entomol. Department of Agriculture, Uganda. 122:169-178. Government Printer, Entebbe, UG. 35 p. Legg, J. P.; French, R.; Rogan, D.; Okao- Okuja, G.; Brown, J. K. 2002. A Herren, H. R.; Neuenschwander, P. 1991. distinct, invasive Bemisia tabaci Biological control of cassava pests (Gennadius) (Hemiptera: in Africa. Ann. Rev. Entomol. Sternorrhyncha: Aleyrodidae) 36:257-283. genotype cluster is associated with the epidemic of severe cassava Hillocks, R. J. 2002. Cassava in Africa. In: mosaic virus disease in Uganda. Hillocks, R. J.; Thresh, J. M.; Mol. Ecol. 11(7):1219-1229. Bellotti, A. C. (eds.). Cassava: Biology, production and utilization. Mound, L. A.; Halsey, S. H. 1978. CAB International, Wallingford, GB. Whitefly of the world. A systematic p. 41-54. catalogue of the Aleyrodidae (Homoptera) with host plant and Hong, Y. G.; Robinson, D. J.; Harrison, natural enemy data. John Wiley, B. D. 1993. Nucleotide sequence Chichester, GB. 340 p. evidence for the occurrence of three distinct whitefly-transmitted Neuenschwander, P. 1994. Control of the geminiviruses in cassava. J. Gen. cassava mealy bug in Africa: Virol. 74:2437-2443. Lessons from a biological control project. Afr. Crop Sci. J. 2:369-384. Jennings, D. L. 1970. Cassava in Africa. Field Crop Abstr. 23:271-278. Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; Baguma, Y.; Ogwal, S.; Semakula, Joyce, R. J. V. 1955. Cotton spraying in G. N.; Acola, G.; Byabakama, B.; the Sudan Gezira. II. Entomological Martin, A. 1997. Cassava mosaic problems arising from spraying. virus disease in Uganda: The FAO Plant Prot. Bull. 3:97-103. current pandemic and approaches to control. Natural Resources Kirkpatrick, T. W. 1931. Leaf-curl of Institute (NRI), Chatham, GB. 65 p. cotton in the Sudan. Bull. Entomol. Res. 21:127-137. Schaefers, G. A.; Terry, E. R. 1976. Insect transmission of sweetpotato Kufferath, H.; Ghesquière, J. 1932. La diseases in Nigeria. Plant Pathol. mosaïque du manioc. Compte- 66:642-645. Rendu Soc. Biol. Belge 109:1146- 1148. Storey, H. H.; Nichols, R. F. W. 1938. Studies on the mosaic of cassava. Legg, J. P. 1994. Bemisia tabaci: The Ann. Appl. Biol. 25:790-806. whitefly vector of cassava mosaic geminiviruses in Africa: An Swanson, M. M.; Harrison, B. D. 1994. ecological perspective. Afr. Crop Sci. Properties, relationships and J. 2:437-448. distribution of cassava mosaic geminiviruses. Trop. Sci. 34:15-25. Legg, J .P. 1999. Emergence, spread and strategies for controlling the Thresh, J. M.; Fargette, D.; Otim-Nape, pandemic of cassava mosaic virus G. W. 1994. The viruses and virus disease in east and central Africa. diseases of cassava in Africa. Afr. Crop Prot. 18:627-637. Crop Sci. J. 2:459-478.

22 Introduction

Thresh, J. M.; Otim-Nape, G. W.; Legg, Yaninek, J. S.; Onzo, A.; Ojo, J. B. 1993. J. P.; Fargette, D. 1997. African Continent-wide releases of cassava mosaic disease: The neotropical phytoseiids against the magnitude of the problem? Afr. J. exotic cassava green mite in Africa. Root Tuber Crops 2:13-19. Exp. Appl. Acarol. 17:145-160.

Thresh, J. M.; Otim-Nape, G. W.; Zhou, X.; Liu, Y.; Calvert, L.; Muñoz, C.; Thankappan, M.; Muniyappa, V. Otim-Nape, G. W.; Robinson, D. J.; 1998. The mosaic diseases of Harrison, B. D. 1997. Evidence that cassava in Africa and India caused DNA-A of a geminivirus associated by whitefly-borne geminiviruses. with severe cassava mosaic disease Rev. Plant Pathol. 77:935-945. in Uganda has arisen by interspecific recombination. J. Gen. Warburg, O. 1894. Die Kulturpflanzen Virol. 78:2101-2111. Usambaras. Mitt. Dtsch. Schutz. 7:131. Zimmerman, A. 1906. Die Krauselkrankheit des Maniok. Yaninek, J. S.; Herren, H. R. 1988. Pflanzer 2:145. Introduction and spread of the cassava green mite Mononychellus tanajoa (Bondar) (Acari: Tetranychidae), an exotic pest in Africa and the search for appropriate control methods: A review. Bull. Entomol. Res. 78:1-13.

23 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.2 Ghana

Anthony Cudjoe*, Joseph Gyamenah* and Braima James**

Introduction and Hosny) was unevenly distributed and less abundant (Sotomey et al., Cassava mosaic disease (CMD) is a 1995). Across ecozones and seasons, major food production constraint in CMD was the most common cassava Africa (Thresh et al., 1994). It is caused pest constraint both in terms of field by begomoviruses, which are vectored data and farmer perceptions. In the by the whitefly Bemisia tabaci field, total plant incidence of CMD was (Gennadius) and can be spread 76% in the dry season and 68% in the through infected planting material. wet and, in 56% of the villages, farmers Yield losses caused by the disease are ranked CMD as the main disease. On higher if vegetative planting materials an ascending 1-5 scale of damage sprout with symptoms (cutting classes, 22% of plants were in class infection) than if initially disease-free 2 (mild), 51% in class 3 (moderate), material becomes infected by B. tabaci 26% in class 4 (severe) and none in following sprouting (current season class 5 (very severe). Previous attempts whitefly-borne infection) (Fauquet and to develop CMD control measures in Fargette, 1990). In Ghana, recent work Ghana have included preliminary on Bemisia whiteflies and CMD has evaluation of local cassava varieties in been carried out under the auspices of relation to their susceptibility to CMD a regional project, Ecologically and laboratory studies to produce Sustainable Cassava Plant Protection virus-free tissue culture materials of (ESCaPP) (Yaninek et al., 1994). local germplasm (Bieler et al., 1996; Unpublished results of ESCaPP Kissiedu et al., 1996). In subsequent diagnostic surveys showed cassava collaborative epidemiology trials, (Manihot esculenta Crantz) to be the whitefly infection was consistently most widely planted arable crop in the lowest in the variety TMS 30001 when country (45% site incidence), followed compared to other improved or local by maize (Zea mays L.) (15%) and yam varieties (Legg et al., 1997; IITA, 1998; (Dioscorea spp.) (5%). James et al., 1998). The role of parasitoids, other natural enemies, Whitefly survey samples consisted genetic heterogeneity in Bemisia largely of B. tabaci. B. afer (Priesner populations and the identity of cassava viruses/viral strains are yet to be * Plant Protection and Regulatory Services investigated in the country. Directorate (PPRSD), Ministry of Food and Agriculture, Ghana. This chapter gives results of a ** International Institute of Tropical Agriculture (IITA), Biological Control Center for Africa, countrywide survey in Ghana, Cotonou, Benin. conducted in November and December

24 Ghana

1997, on 80 farms. These included symptoms in various local names. 25 sites in each of the rainforest and Some of these, for example, bankye transition forest, 10 sites in the kwata and kwata, described the “coastal” savannah and 20 sites in the appearance of CMD symptoms. Such wet savannah (Figure 1). indigenous knowledge will facilitate farmer participatory learning and research activities. Bemisia whitefly Burkina Faso abundance averaged 1.3 adults per cassava shoot tip in the rainforest, Northern guinea savannah 1.0 in transition forest, 11.8 in coastal savannah and 0.3 in wet savannah (see also Figure 2 in Chapter 1.14, this Southern guinea savannah volume). Whitefly mean abundance Togo (logarithm transformed counts) was significantly higher in the coastal Côte Derived d’Ivoire savannah savannah than in each of the other ecozones (t > 12.1; P < 0.05). The Ghana natural enemy fauna associated with the whiteflies included the hymenopteran parasitoid Encarsia sophia (Girault and Dodd), which was Humid forest reared from Bemisia mummies at four Coastal derived survey sites. savannah

Increased Socio-Economic Understanding Figure 1. Areas surveyed for whitefly incidence and cassava mosaic disease in Ghana. Cassava production was characterized by small landholdings, typically less than 1 ha in size, and a preponderance Increased Biological of local landraces. The intensity of Understanding cassava farming decreased along the rainforest-savannah axis. The five most B. tabaci, B. afer and Trialeurodes profitable crops listed by farmers were vaporariorum (Westwood) were cassava (45%), yam (18%), maize identified from adult whitefly samples (11%), cowpea (Vigna unguiculata [L.] collected from cassava in Ghana. Walp.) (6%) and tomato (Lycopersicon Whilst B. tabaci occurred at all survey esculentum Mill.) (5%). These other sites, B. afer was recorded from only crops were planted in association with, 3.8% of the sites and comprised 1.9% or near to, cassava fields but cassava of the 151 specimens determined. was the most common crop (40% site T. vaporariorum was collected from only incidence) planted in nearby fields. one location. Generally, farmers were Among the associated crops, both unaware of Bemisia whiteflies and did cowpea and tomato are known host not recognize them by specific local plants of B. tabaci. However, names; only 2% of them knew that the experimental data from other countries insects caused serious problems to in Africa show that the B. tabaci cassava. By contrast, 54% of farmers biotype that occurs on cassava is knew that CMD caused serious largely restricted to that crop (Burban problems and described the disease et al., 1992; Legg et al., 1994).

25 Whiteflies and Whitefly-borne Viruses in the Tropics

Priority cassava pests listed by to pest-induced losses. In view of the farmers were vertebrates (16%), weeds higher proportion of plants in the (13%), termites and weevils (10% each) moderate and severe damage categories and CMD (6%). Overall, CMD incidence in the transition forest, significant root was mainly attributable to cutting yield losses could be expected in this infection (Figure 2). The incidence of ecozone. However, yield loss estimates cutting infection averaged 60% in the provided by farmers appeared to be rainforest, 55% in the transition forest, unrealistically high. For example, about 63% in the coastal savannah and 51% 26% of farmers estimated losses at in the wet savannah. The incidence of 50% and 20% of farmers at 75%. Such whitefly infection, transformed to allow loss estimates would need to be for the effect of multiple infection validated, especially since 91% of the (Gregory, 1948), was significantly farmers reported that they sell their higher in the coastal savannah and cassava harvest. rainforest than in the transition forest or wet savannah (t > 2.9; P < 0.05). Whitefly abundance showed no 100 90 correlation with the incidence of 80 whitefly infection. In terms of reducing 70 CMD incidence, the transition forest 60 50 and wet savannah would seem to 40 provide better sites for multiplication of 30 clean planting material. 20 10

different CMD severities 0 Percentage of plants with Slight Moderate Serious damage damage damage 100 Rainforest 80 Transition forest Wet savannah Coastal savannah 60

40 Figure 3. Cassava mosaic disease (CMD) damage severity in the ecozones of 20 Ghana. Percentage of plants with CMD symptoms 0 Farmers’ opinion was almost

Wet equally divided on whether or not CMD forest Coastal savannah savannah Rainforest Transition was becoming more severe, and a Whitefly infection Cutting infection sizeable proportion (34%) considered that the disease was not a yearly Figure 2. Cassava mosaic disease (CMD) incidence and source of infection in occurrence. Extension and training the ecozones of Ghana. support for whitefly/CMD problems appeared low, since 89% of farmers reported receiving no technical Across ecozones, 24%-32% of information or assistance with these plants showed no CMD symptoms. problems. In view of this, it is not About 10% of the plants showed surprising that only 1% of farmers serious (score 4) to severe (score 5) practiced any deliberate whitefly/CMD damage symptoms (Figure 3). The control. Local varieties predominated plants were 3-6 months old, a growth (54% site incidence) and 18% of the stage during which storage root farmers ranked Busuminsia as their formation and development is initiated single most preferred variety, followed and root yield is particularly vulnerable by three other local varieties, Santom

26 Ghana

(6.2%), Bakentenma (5%) and Kokoo rogue diseased plants, mostly (55% of (5%). Adoption of improved varieties farmers) within the first month of was low: only 11% of farmers planted planting. TMS 30572 (originating from the International Institute of Tropical Agriculture [IITA]) and “Agric” varieties Conclusions (a locally adapted material of IITA origin) recommended by the national This study has increased field-level extension service on the basis of their understanding of whiteflies and CMD CMD resistance. Agronomic features incidence, and provides a strong basis were more important criteria for for the development of ecologically selecting planting material than were sound management practices for disease-specific characteristics. whitefly/CMD problems in Ghana. The incidence of CMD infection by B. tabaci Where disease-specific selection was low and the disease appears to be criteria were used, only 10% of the spread mainly through infected farmers reported that the method was cuttings. The preponderance of local effective, while 9% noticed that the cassava varieties over proven CMD- “clean” material became diseased after resistant improved varieties contributes planting. Since the single most to the incidence of the disease in the important primary source of planting country. Two key issues that need to be material was farmers’ own fields addressed are the quantification of (Figure 4), adoption of appropriate yield losses in the preferred local procedures for selecting planting cultivars and, associated with this, an material would significantly reduce assessment of why CMD-resistant recycling of cassava mosaic viruses. varieties have not been adopted widely. About 60% of farmers reported that they encountered shortage of planting Assuming that yield losses warrant material in some years. Shortages like the use of CMD control measures, these, coupled with the observed high farmer participatory learning activities incidence of cutting infection, imply will promote farm-level selection of that rapid multiplication schemes parent planting material that is CMD would be required to ensure timely symptom free for making cuttings. availability of clean and healthy Related to this, farmer-led rapid planting materials at farm level. No multiplication schemes would be farmer reported pesticide use as a required to increase the availability and control method against the whiteflies or adoption of CMD-resistant varieties, the disease but about 61% of farmers provided that evaluations linked to such schemes indicate that some of these varieties are acceptable to farmers. Research should also evaluate Own farm farmers’ preferred local varieties as Neighbour possible sources of CMD resistance in

Extension breeding programs since this approach has provided important new impetus to Market resistance breeding programs for 010203040506070 cassava elsewhere (Dixon et al., 1992). Relative frequency (%) of planting material source Little is known about temporal and Figure 4. Farmers’ sources of cassava spatial changes in the abundance of planting material in Ghana. B. tabaci in Ghana, particularly in

27 Whiteflies and Whitefly-borne Viruses in the Tropics relation to its ability to transmit Burban, C.; Fishpool, L. D. C.; Fauquet, cassava mosaic viruses over the C.; Fargette, D.; Thouvenel, J. -C. growing period of the crop. Field 1992. Host associated biotypes within West African populations of studies therefore might be necessary to the whitefly Bemisia tabaci (Genn.) quantify such changes, assess whether (Hom., Aleyrodidae). J. Appl. parasitism by E. sophia has any Entomol. 113:416-423. significant impact on the insect’s abundance and, in epidemiology trials, Dixon, A. G. O.; Asiedu, R.; Hahn, S. K. to ascertain the ability of B. tabaci to 1992. Cassava germplasm spread cassava mosaic viruses to a enhancement at IITA. In: Akoroda, M. O.; Arene, O. B. (eds.). range of cassava varieties. Promotion of root crop-based industries: An incentive for research National research capacity needs to and development. Procs. 4th be enhanced in order to allow project Triennial Symposium of the partners to undertake further research International Society for Tropical on the whitefly/CMD issues raised Root Crops-Africa Branch (ISTRC- here. In this regard, training of national AB), 5-8 December 1989, Kinshasa, CD. p. 83-87. program partners in appropriate areas would strengthen the collaborative Fauquet, C.; Fargette, D. 1990. African linkages initiated in the pilot phase of cassava mosaic virus: Etiology, the project and provide a strong base epidemiology, and control. Plant for providing sustainable technical Dis. 74:404-411. support to cassava farmers. Key areas for strengthened national research Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. capacity include post-graduate degree Biol. 35:412-417. training in B. tabaci bionomics and cassava mosaic virology, and field- IITA (International Institute of Tropical based training of technicians and Agriculture). 1998. Ecologically extension workers in participatory Sustainable Cassava Plant processes and facilitation. Technician Protection (ESCaPP) in South and extension training will increase America and Africa. Africa: Project findings and recommendations. scientific literacy in farming Ibadan, NG. communities and foster action research and learning with farmers. This will James, B. D.; Legg, J. P.; Gbaguidi, J. B.; improve farmers’ currently poor access Annang, D.; Asiabaka, C. C.; to information, will help them Cudjoe, A. R.; Echendu, T. N. C.; understand the causes and nature of Maroya, N. G.; Ntonifor, N.; Ogbe, the disease problem and will highlight F.; Salawu, R. A.; Tumanteh, J. A. 1988. Understanding farm-level effective cultural control options such epidemiology of cassava mosaic as planting material selection and disease. In: Abstracts, 7th Triennial roguing. Symposium of the International Society for Tropical Root Crops- Africa Branch (ISTRC-AB), 11-17 References October 1988, Cotonou, BJ. p. 122.

Bieler, P.; Ekanayake, I. J.; Dossou, R.; Kissiedu, A. F. K.; Asare Bediako, A.; Sanni, A. O.; Ahihou, K.; Alhassan, Afuakwa, J. J.; Okai, E.; Missah, A. A. Y.; Ahiabu, R.; Yaninek, J. S. 1996. National Root and Tuber 1996. Production of high quality Crops Improvement Project cassava planting materials: A (NRTCIP), Ghana-highlights of collaborative effort. Trop. Root activities and achievements 1992- Tuber Crops Bull. 9(1):1-3. 93. Trop. Root Tuber Crops Bull., Special Edn. March 1996. p. 3-5.

28 Ghana

Legg, J. P.; James, B. D. 1998. Tackling Sotomey, M.; James, B. D.; Gbaguidi, B. whitefly and whitefly-borne virus 1995. Les espèces de Bemisia sur le constraints to cassava and manioc en Afrique occidentale et sweetpotato in Africa: Aims, centrale. In: 11th Meeting and approach, and progress of the Scientific Conference of the African CGIAR whitefly IPM project. In: 7th Association of Insect Scientists, Triennial Symposium of the 6-11 August 1995, Yamoussoukro, International Society for Tropical CI. p. 4. Root Crops-Africa Branch (ISTRC- AB), 11-17 October 1988, Cotonou, Thresh, J. M.; Fishpool, L. D. C.; Otim- BJ. p. 64. Nape, G. W.; Fargette, D. 1994. African cassava mosaic virus Legg, J. P.; Gibson, R. W.; Otim-Nape, G. disease: An under-estimated and W. 1994. Genetic polymorphism unsolved problem. Trop. Sci. amongst Ugandan populations of 34:3-14. Bemisia tabaci (Gennadius) (Homoptera: Aleyrodidae), vector of Yaninek, J. S.; James, B. D.; Bieler, P. African cassava mosaic geminivirus. 1994. Ecologically Sustainable Trop. Sci. 34:73-81. Cassava Plant Protection (ESCaPP): A model for environmentally sound Legg, J. P.; James, B. D.; Cudjoe, A.; pest management in Africa. Afr. Saizonou, S.; Gbaguidi, B.; Ogbe, Crop Sci. J. 2:553-562. F.; Ntonifor, N.; Ogwal, S.; Thresh, J.; Hughes, J. 1997. A regional collaborative approach to the study of ACMD epidemiology in sub- Saharan Africa. Afr. Crop Sci. J. 3:1021-1033.

29 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.3 Benin

Brice Gbaguidi*, Braima James* and Symphorien Saizonou**

Introduction (indicating “slight damage” on a 1-5 scale) across ecozones and Preliminary research on cassava pests seasons. During the surveys, in focus and diseases in Benin was carried out group interviews at village level, as part of a regional project for farmers at 68% of survey sites rated Ecologically Sustainable Cassava Plant CMD as the main disease problem. In Protection (ESCaPP) (Yaninek et al., subsequent preliminary CMD 1994). Unpublished results of ESCaPP epidemiology trials (Legg et al., 1997; diagnostic surveys show cassava IITA, 1998; James et al., 1998), current (Manihot esculenta Crantz) to be the season whitefly infection was predominant arable crop (87% site consistently lowest in the variety incidence) in the country and the next TMS 30001 when compared with other most common crops to be maize (Zea improved or local varieties. In a related mays L.) with 30% site incidence and population dynamics study, the cowpea (Vigna unguiculata [L.] Walp.) parasitoid Encarsia sophia (Girault and with 26%. Summaries of plant Dodd) was identified from Bemisia protection aspects of the survey show mummies at each of six trial sites in cassava mosaic disease (CMD) caused the transition forest, and wet and dry by cassava mosaic begomoviruses savannahs (James and Gbaguidi, (CMBs) and vectored by the whitefly unpublished data). Genetic Bemisia tabaci (Gennadius) to be the heterogeneity in B. tabaci populations, most common biotic constraint on which is known to influence CMD cassava in the country, with total plant epidemiology (Burban et al., 1992; Legg incidence of 49% in the dry season and et al., 1994), and the identities of 57% in the wet. Survey samples of cassava viruses and virus strains are red-eyed whitefly nymphs comprised yet to be investigated in the country. B. tabaci and B. afer (Priesner and Hosny) although the latter was less This chapter reports the results of abundant, occurring in only 4% of the a countrywide diagnostic survey of samples (Sotomey et al., 1995). The whiteflies and CMD in Benin, CMD damage score was low, the conducted in November and December average falling within class 2 1997, on 60 farms distributed in the transition forest (28 sites), dry savannah (19 sites) and wet savannah * International Institute of Tropical Agriculture (13 sites) (Figure 1). Farmer (IITA), Biological Control Center for Africa, respondents in the survey were mostly Cotonou, Benin. ** Service Protection des Végétaux et Contrôle men (93%) and landowners with over Phytosanitaire, Porto Novo, Benin. 5 years of cassava farming experience.

30 Benin

ekpo, goudou and kpanro refer specifically to the resemblance of CMD symptoms to those of leprosy. This indigenous knowledge will greatly facilitate farmer participatory research and learning activities—activities that Northern guinea seem especially important given that a savannah high proportion of farmers did not Southern guinea know that whiteflies (60% farmers) and savannah CMD (50% farmers) caused serious problems to cassava. Whilst B. tabaci occurred at all 60 survey sites, B. afer was recorded in 50% of the sites and Benin comprised 15% of the 212 specimens determined. Whitefly abundance was Nigeria low and averaged 3.2 adults per Togo cassava shoot tip in the transition forest, 0.5 in the wet savannah and Derived 0.9 in the dry savannah. Whitefly mean savannah abundance (logarithm transformed counts) was significantly higher in the transition forest than in either wet or dry savannah (t > 5.04; P < 0.05). The Forest parasitoids E. sophia and Eretmocerus sp. were identified in natural enemy collections. The presence of E. sophia Figure 1. Areas surveyed for whitefly was confirmed at 15 sites and incidence and cassava mosaic Eretmocerus sp. at only one site. disease in Benin. Diseased whiteflies were not observed in the field and hence no Based on the survey results, the entomopathogens were collected. chapter identifies areas requiring more Biotypes of cassava viruses identified research attention to further our are reported elsewhere (Chapter 1.11, understanding of CMD in the country this volume). and to serve as a basis for developing and implementing an integrated approach to managing this disease. Increased Socio-Economic Understanding

Increased Biological Across ecozones, cassava was typically Understanding produced on farms of less than 0.25 ha. The improved variety “Agric”, The co-existence of B. tabaci and B. originating from the International afer on cassava in Benin was Institute of Tropical Agriculture (IITA), confirmed by the determination of was grown by 13% of farmers. The these species from adult specimens. In local variety Odounbo, grown by 7% of various local languages, farmers farmers, was the next most popular recognized whiteflies simply as variety, while the remainder comprised “insects”. On the other hand, among a wide diversity of local cultivars. Fifty- the various local names used to eight percent of farmers ranked describe CMD, the names edjekpo, cassava as their most profitable crop,

31 Whiteflies and Whitefly-borne Viruses in the Tropics compared to 33% for maize, 10% each (t > 2.06; P < 0.05). Across ecozones, for cowpea and groundnut (Arachis whitefly abundance showed no hypogaea L.) and 5% for cotton correlation with the incidence of (Gossypium hirsutum L.). These crops whitefly infection. The pattern of were planted also in association with damage was similar across ecozones, cassava or near to cassava fields. with most plants showing no CMD Cowpea, groundnut and cotton are symptoms and, of those visibly infected known crop host plants of B. tabaci, with the disease, only a very small although experimental data from other proportion showing moderate or severe countries in Africa show that the damage (Figure 3). This is important in B. tabaci biotype occurring on cassava the context of the age of cassava farms is more or less restricted to that crop sampled. The plants were 3-6 months (Burban et al., 1992; Legg et al., 1994). old at the time of the survey, an age at Cassava was also the most common which storage roots form and begin to crop planted in nearby fields (38% of develop and root yield is particularly all instances) and the intensity of vulnerable to pest-induced losses. cassava farming gradually decreased None of the farmers reported total loss from the transition forest (2.4 nearby of their cassava to CMD, probably a fields) to wet savannah (1.9 nearby reflection of the mild disease symptoms fields) and dry savannah (0.2 nearby observed. However, 25% of farmers fields). attributed losses of at least 25% of the yield to the disease. Whilst this high Average CMD incidence ranged loss estimate seems to conflict with the from 25% to 49% across ecozones low damage severity observed, 58% of (Figure 2), with infected cuttings farmers indicated that CMD is providing the more important source of becoming more severe. infection. The incidence of cutting infection did not differ significantly between ecozones. The incidence of 100 whitefly infection, transformed to allow 90 for the effect of multiple infection 80 70 (Gregory, 1948), differed significantly 60 across ecozones and was higher in 50 both the transition forest and wet 40 30 savannah than in the dry savannah 20 10

different CMD severities 0 Percentage of plants with Slight Moderate Serious damage damage damage 60 Transition forest Wet savannah 40 Dry savannah

Figure 3. Cassava mosaic disease (CMD) 20 damage severity in the ecozones of Benin. Percentage of plants with CMD symptoms 0 Transition Wet Dry Most farmers (78%) undertook no forest savannah savannah specific control measures against CMD Whitefly infection Cutting infection or whiteflies, and less than 10% had received technical information or Figure 2. Cassava mosaic disease (CMD) incidence and source of infection in assistance on them. Among farmers the ecozones of Benin. who attempted to control the whitefly

32 Benin or disease problem, the main methods Conclusions used were weeding and application of wood ash. Even though no farmer In Africa, cassava root yield losses due reported the use of resistant varieties to CMD have been estimated at as a whitefly/disease control method, 28%-40% annually (Thresh et al., further questioning revealed that 8% of 1994). In Benin, there appears to be a farmers listed “Agric” and 5% listed contradiction between field data and Odounbo as varieties they had planted farmers’ perception of the CMD on the basis of recommendations that problem. The research data indicate they were tolerant to CMD. Twenty only a very small proportion of cassava percent of these farmers indicated that plants with moderate or severe disease the better performance of these damage but 25% of the farmers varieties was mainly because of better attribute high losses to the disease. In sprouting and establishment of the such cases, farmer participatory trials cutting. Agronomic features were more would help farmers and researchers important than pest/disease resistance jointly to better understand the as criteria for the selection of planting relationship between disease incidence material. About 12% of the selection and yield, and to validate loss criteria related to disease-free stems, estimates. The studies would help health of stems and disease resistance, establish whether losses to CMD really whilst 23% related to size, maturity, are significant in Benin and, if so, better yield, colour and size of the convince farmers of the need to adopt parent stem. integrated management practices against the disease. Such an approach Where disease-specific criteria were will provide for action learning by used for selecting planting material, farmers to improve their access to only 10% of the farmers reported that information and understanding of the the method was effective and 7% causes and nature of the disease noticed that the “clean” material problem. For example, cutting infection became diseased after planting. was found to be much more important Because the single most frequently than Bemisia whiteflies in the spread of cited sources of planting material were CMD, suggesting that a participatory farmers’ own fields (30%) and learning and extension effort focusing neighbours’ fields (22%), adoption of on planting material selection and appropriate planting material selection sanitation and roguing could have a would be expected to reduce recycling positive impact. of CMBs considerably. Whilst no farmer indicated roguing as a specific Generally, training of national CMD control method, 7% of them program partners would strengthen the conducted roguing to prevent and collaborative links established in the reduce disease and 2% because of poor pilot phase of the project and further plant growth; 7% of the farmers enhance national capacity to reported some impact from this control undertake priority CMD and whitefly measure. However, the current high research activities. For example, field incidence of cutting infection would studies on the population dynamics of make roguing very labor intensive and B. tabaci would be needed to unattractive for adoption, unless understand temporal and spatial carried out in combination with the changes in the insect’s abundance in selection of CMD-free planting relation to its ability to spread CMD. material. Pesticide use was rare. There is also the need to quantify the effectiveness of parasitism by E. sophia,

33 Whiteflies and Whitefly-borne Viruses in the Tropics particularly as a possible factor Legg, J. P.; Gibson, R. W.; Otim-Nape, G. influencing the importance of the W. 1994. Genetic polymorphism vector in spreading CMD. In view of the amongst Ugandan populations of Bemisia tabaci (Gennadius) low abundance of B. afer in the (Homoptera: Aleyrodidae), vector of country, it would not be appropriate to African cassava mosaic geminivirus. commit further resources to Trop. Sci. 34:73-81. investigating its role in CMD epidemiology. Legg, J.; James, B. D.; Cudjoe, A.; Saizonou, S.; Gbaguidi, B.; Ogbe, F.; Ntonifor, N.; Ogwal, S.; Thresh, J.; Hughes, J. 1997. A regional References collaborative approach to the study of ACMD epidemiology in sub- Burban, C.; Fishpool, L. D. C.; Fauquet, Saharan Africa. Afr. Crop Sci. J. C.; Fargette, D.; Thouvenel, J. -C. 3:1021-1033. 1992. Host associated biotypes within West African populations of Sotomey, M.; James, B. D.; Gbaguidi, B. the whitefly Bemisia tabaci (Genn.) 1995. Les espèces de Bemisia sur le (Hom., Aleyrodidae). J. Appl. manioc en Afrique occidentale et Entomol. 113:416-423. centrale. In: 11th meeting and scientific conference of the African Gregory, P. H. 1948. The multiple Association of Insect Scientists, infection transformation. Ann. Appl. 6-11 August 1995, Yamoussoukro, Biol. 35:412-417. CI. p. 4.

IITA (International Institute of Tropical Thresh, J. M.; Fishpool, L. D. C.; Otim- Agriculture). 1998. Ecologically Nape, G. W.; Fargette, D. 1994. Sustainable Cassava Plant African cassava mosaic virus Protection (ESCaPP) in South disease: An under-estimated and America and Africa. Africa: Project unsolved problem. Trop. Sci. findings and recommendations. 34:3-14. Ibadan, NG. Yaninek, J. S.; James, B. D.; Bieler, P. James, B. D.; Legg, J. P.; Gbaguidi, J. B.; 1994. Ecologically Sustainable Annang, D.; Asiabaka, C. C.; Cassava Plant Protection (ESCaPP): Cudjoe, A. R.; Echendu, T. N. C.; A model for environmentally sound Maroya, N. G.; Ntonifor, N.; Ogbe, pest management in Africa. Afr. F.; Salawu, R. A.; Tumanteh, J. A. Crop Sci. J. 2:553-562. 1998. Understanding farm-level epidemiology of cassava mosaic disease. In: Abstracts, 7th Triennial Symposium of the International Society for Tropical Root Crops- Africa Branch (ISTRC-AB), 11-17 October 1988, Cotonou, BJ. p. 122.

34 Nigeria

CHAPTER 1.4 Nigeria

Nnamdi C. Echendu*, Joseph B. Ojo**, Braima James*** and Brice Gbaguidi***

Introduction pest constraint on cassava in Nigeria with, on average, 85% of plants showing Cassava (Manihot esculenta Crantz) is symptoms in the dry season and 79% in an important staple food crop in sub- the wet (IITA, 1997). The surveys Saharan Africa. In Nigeria, indicated the co-existence of B. tabaci unpublished results of diagnostic and B. afer (Priesner & Hosny) on surveys by a project for Ecologically cassava in Nigeria, although B. afer was Sustainable Cassava Plant Protection less abundant and widespread than (ESCaPP) (Yaninek et al., 1994) show B. tabaci (Sotomey et al., 1995). cassava to be the most important crop in the country, followed by yam In Nigeria, studies are continuing to (Dioscorea spp.) and maize (Zea mays examine the begomoviruses infecting L.). Cassava has the potential to yield cassava and their variability (Ogbe 60 t/ha of storage roots (Nweke and et al., 1998). These have recorded the Lynam, 1997). However, average yields presence of both African cassava mosaic in farmers’ fields are usually less than virus (ACMV) and East African cassava 10 t/ha (FAO, 1989). The shortfall is mosaic virus (EACMV), although earlier associated largely with pest and less comprehensive studies had disease problems, of which cassava suggested the occurrence of only ACMV mosaic disease (CMD) is one of the in Nigeria (Swanson and Harrison, most important. In sub-Saharan Africa 1994). Genetic heterogeneity in B. tabaci as a whole, losses to CMD have been populations, which is known to estimated at 15-27 million tons, influence CMD epidemiology (Burban et representing 15%-24% of total al., 1992; Legg et al., 1994), has yet to production (Thresh et al., 1997). The be investigated in the country. In disease is caused by begomoviruses preliminary CMD epidemiology trials, transmitted by the whitefly Bemisia current season whitefly infection was tabaci (Gennadius) and is spread also consistently lowest in the variety by man through planting of infected TMS 30001 when compared to other planting materials (Hahn et al., 1981). improved or local varieties (Legg et al., The ESCaPP survey results showed 1997; IITA, 1998; James et al., 1998). CMD to be the single most common The development and deployment of resistant varieties has been the main approach to CMD management pursued * National Root Crops Research Institute in Nigeria (Hahn et al., 1980; 1981). (NRCRI), Umudike, Nigeria. Little attention has been given to the ** International Institute of Tropical potential for using natural enemies to Agriculture (IITA), Ibadan, Nigeria. *** IITA, Biological Control Center for Africa, manage the whitefly vector, although Cotonou, Benin. Jerath (1967) suggested that the 35 Whiteflies and Whitefly-borne Viruses in the Tropics parasitoid Encarsia strenua (Silvestri) were found in samples collected. Some and a predatory mite, Typhlodromus sp., farmers recognized whiteflies and 17.5% were possible biocontrol agents. of farmers knew that the insects caused a serious problem to cassava. In various This report presents results of a local languages, farmers recognized countrywide survey of Bemisia whiteflies Bemisia whiteflies simply as “insects”. and CMD conducted in November and About 24% of farmers knew that CMD December 1997, on 80 farms caused serious problems and farmers in distributed across the main cassava- different localities described the disease producing agro-ecological zones of in various local languages as akpa, Nigeria (Figure 1). Sites were sampled in ekikang, ikwukwo, kpuojuju and rainforest (25), transition forest (25), wet nkpuk—mostly terms that refer to the savannah (20) and dry savannah (10) appearance of the disease symptoms. zones, and cassava plantings selected Bemisia abundance was less than three were 3 to 4 months old. This chapter adults per cassava shoot tip (defined as summarizes the results of the survey the terminal shoot with the first five and identifies areas requiring open leaves) in each of the ecozones. subsequent attention in order to further our understanding of the disease and to Five species of hymenopterous serve as a basis for developing and parasitoids of Bemisia were recorded: implementing the integrated Encarsia sophia (Girault and Dodd) management of CMD in Nigeria. comprising 93.7% of samples, Eretmocerus sp., 3.5%, Encarsia sp. (luteola group), 0.7%, Encarsia lutea (Masi), 1.2% and Encarsia mineoi Niger Viggiani, 0.8%. Potential whitefly predators recorded were Cheilomenes Northern guinea sulphurea (Olivier) and phytoseiid mites. Benin savannah Diseased whiteflies were not observed in Southern guinea the field and entomopathogens therefore savannah Nigeria were not collected. The diversity in Derived savannah cassava viruses identified is reported in Chapter 1.11 (this volume). Other Humid forest cassava pests listed by farmers were Cameroon grasshoppers, reported by 15% of farmers, termites by 6.3%, and vertebrates by 2.5%. Figure 1. Areas surveyed for whitefly incidence and cassava mosaic disease in Nigeria. Increased Socio-Economic Understanding Increased Biological Understanding Cassava farm sizes averaged 0.38 hectares in the rain and transition The insect vector of CMD, B. tabaci, forests, 0.56 hectares in the wet occurred at all survey sites, while a savannahs and 0.10 hectares in the dry second whitefly species, Trialeurodes savannahs. Both local and improved vaporariorum (Westwood), hitherto varieties were recorded in farmers’ unreported on cassava, was identified fields, although local varieties from nymphs collected from one location predominated. The most widely in the country. No examples of B. afer occurring improved varieties were

36 Nigeria

TMS 30572 (18.8%) and TMS 30555 vulnerable to pest-induced losses. Very (2.5%). There was a marked difference few farmers (1.2%) reported total yield in the incidence of improved varieties loss in the first planting season but between ecozones: they were cultivated 18.8% of farmers attributed losses of at in 80% of sampled fields in the least one quarter of their yield to the transition forest but occurred in only disease. However, about 70% of the about 30% of fields in the rainforest farmers provided no yield loss and wet savannah and were not estimates for the first or second crop. recorded at all in the dry savannah. There was a direct negative correlation between the frequency of cultivation of 100 improved varieties and CMD incidence. 80 Farmers ranked cassava as their most 60 profitable crop followed by maize, yam, sorghum (Sorghum bicolor [L.] Moench) 40 and cowpea (Vigna unguiculata [L.] 20 Walp.), in descending order. Of the Percentage of plants crops planted in association with with CMD symptoms 0 cassava or in adjacent fields, cowpea is Dry a known host plant of B. tabaci. Wet forest savannah savannah

Experimental data from other countries Transition Rainforest in Africa, however, show that the B. Whitefly infection Cutting infection tabaci biotype occurring on cassava is more or less restricted to that crop Figure 2. Cassava mosaic disease (CMD) incidence and source of infection in (Burban et al., 1992; Legg et al., 1994). the ecozones of Nigeria.

Average CMD incidence ranged from 45% to 83% across ecozones, with 100 90 cuttings providing by far the more 80 important source of infection (43% to 70 60 83%) (Figure 2). The incidence of 50 whitefly infection, transformed to allow 40 for the effect of multiple infection 30 20 (Gregory, 1948), was low in each of the 10

different CMD severities 0 ecozones. Among plants with disease, Percentage of plants with Slight Moderate Serious slight damage symptoms (score 2 on a damage damage damage five-point scale from no damage to severe damage) predominated Rainforest Transition forest (Figure 3). Plant damage severity varied Wet savannah Dry savannah little by ecozone. Moderate damage Figure 3. Cassava mosaic disease (CMD) symptoms (score 3) were more damage severity in the ecozones of pronounced in the rainforest (30% of Nigeria. plants) and wet savannah (20% of plants) than in the other ecozones, Nearly all farmers reported that whilst serious (score 4) to severe CMD occurred every year, while slightly (score 5) damage symptoms were more than half of them confirmed that negligible (Figure 3). Plants were less CMD was most severe during the early than 4 months old at the time of the growth stages of cassava. More than surveys, a growth stage during which three-quarters of farmers (77.5%) storage root formation and development reported that they had received no is initiated and root yield is particularly technical information or assistance on

37 Whiteflies and Whitefly-borne Viruses in the Tropics

CMD or whiteflies. Among farmers who based on the relatively mild symptoms attempted to manage the whitefly/ that were observed in the survey. There disease problem, the main practices is an obvious need, however, to obtain used were selection of planting yield loss data for the commonly grown materials, application of wood ash, varieties in order to make an accurate roguing and proper weeding. Agronomic assessment of economic impact. In view features were more important than of the overwhelming importance of pest/disease resistance as criteria for cutting infection, as compared with the selection of planting material. Since vector-borne infection, in the spread of the most frequently cited sources of CMD, research and extension efforts on planting material were neighbours’ and assuring the health of planting material farmers’ own fields (Figure 4), adoption should be prioritized. The farmers’ of appropriate strategies for selecting generally high level of awareness of planting material would be expected to CMD and their willingness to rogue reduce the incidence of CMD diseased plants to prevent or slow its appreciably. About 53% of farmers spread could provide a good basis for rogued CMD affected plants, usually establishing participatory experiments within 3 months of planting. Most to help farmers multiply clean planting farmers (61%), however, rated roguing material of their preferred varieties. as partially effective, with only 4% Action learning activities, involving considering it to be highly effective. In farmers and extension agents, could the choice of varieties, 25% of farmers help improve their access to information cited “Agric” as the single most and their understanding of the causes commonly planted variety and indicated and nature of the disease problem. This that the choice was based on will provide opportunities to promote recommendations that it was resistant cultural control options such as to CMD. Other CMD-resistant varieties appropriate methods of selecting commonly recorded were TMS 30572, planting material, sanitation and TMS 30001 and TMS 30555. No farmer roguing. Additionally, training of used pesticides against CMD and fewer national program partners would than 2% of farmers used pesticides strengthen national capacity to against whiteflies. investigate and promote the sustainable management of whiteflies and CMD.

Neighbour Acknowledgements Own farm Market The assistance of Dr. M. A. Bob of the Research International Center of Insect Physiology Extension and Ecology in the identification of whitefly species and natural enemies is 0 5 10 15 20 25 30 35 40 recognized. Relative frequency (%) of planting material source Figure 4. Farmers’ sources of cassava planting material in Nigeria. References

Burban, C.; Fishpool, L. D. C.; Fauquet, Conclusions C.; Fargette, D.; Thouvenel, J. -C. 1992. Host associated biotypes within West African populations of In Nigeria, whilst incidence of CMD is the whitefly Bemisia tabaci (Genn.) high, the economic impact of the (Hom., Aleyrodidae). J. Appl. disease is believed to be only moderate, Entomol. 113:416-423.

38 Nigeria

FAO (Food and Agriculture Organization Legg, J. P.; Gibson, R. W.; Otim-Nape, G. of the United Nations). 1989. W. 1994. Genetic polymorphism Production year book. Rome, IT. amongst Ugandan populations of Bemisia tabaci (Gennadius) Gregory, P. H. 1948. The multiple (Homoptera: Aleyrodidae), vector of infection transformation. Ann. Appl. African cassava mosaic geminivirus. Biol. 35:412-417. Trop. Sci. 34:73-81.

Hahn, S. K.; Terry, E. R.; Leuschner, K. Legg, J. P.; James, B. D.; Cudjoe, A.; 1980. Breeding cassava for Saizonou, S.; Gbaguidi, B.; Ogbe, F.; resistance to cassava mosaic Ntonifor, N.; Ogwal, S.; Thresh, J.; disease. Euphytica 29:673-683. Hughes, J. 1997. A regional collaborative approach to the study of Hahn, S. K.; Terry, E. R.; Leuschner, K.; ACMD epidemiology in sub-Saharan Singh, T. P. 1981. Cassava Africa. Afr. Crop Sci. J. 3:1021-1033. improvement strategies for resistance to major economic Nweke, F. I.; Lynam, J. K. 1997. Cassava diseases and pests in Africa. In: in Africa. Afr. J. Root Tuber Crops Procs. Triennial Root Crops 2:10-13. Symposium, 1980, International Development Research Centre Ogbe, F. O.; Atiri, G. I.; Thottappilly, G.; (IDRC), Ottawa, CA. p. 279. Dixon, A. G. O.; Mignouna, H. D.; Quin, F. M. 1998. Evidence of double IITA (International Institute of Tropical infection and random occurrence of Agriculture). 1997. Ecologically cassava begomoviruses in sub- Sustainable Cassava Plant Saharan Africa. In: Procs. 7th Protection (ESCaPP) diagnostic Triennial Symposium of the survey results: Dry and wet season International Society for Tropical plant protection summaries. Root Crops-Africa Branch (ISTRC- Ibadan, NG. AB), 11-17 October 1988, Cotonou, BJ. p. 524-529. IITA (International Institute of Tropical Agriculture). 1998. Ecologically Sotomey, M.; James, B. D.; Gbaguidi, B. Sustainable Cassava Plant 1995. Les espèces de Bemisia sur le Protection (ESCaPP) in South manioc en Afrique occidentale et America and Africa. Africa: Project centrale. In: 11th Meeting and findings and recommendations. Scientific Conference of the African Ibadan, NG. Association of Insect Scientists, 6-11 August 1995, Yamoussoukro, James, B. D.; Legg, J. P.; Gbaguidi, J. B.; CI. p. 4. Annang, D.; Asiabaka, C. C.; Cudjoe, A. R.; Echendu, T. N. C.; Swanson, M. M.; Harrison, B. D. 1994. Maroya, N. G.; Ntonifor, N.; Ogbe, Properties, relationships and F.; Salawu, R. A.; Tumanteh, J. A. distribution of cassava mosaic 1998. Understanding farm-level geminiviruses. Trop. Sci. 34:15-25. epidemiology of cassava mosaic disease. In: Abstracts, 7th Triennial Thresh, J. M.; Otim-Nape, G. W.; Legg, J. Symposium of the International P.; Fargette, D. 1997. African cassava Society for Tropical Root Crops- mosaic disease: The magnitude of the Africa Branch (ISTRC-AB), problem? Afr. J. Root Tuber Crops 11-17 October 1998, Cotonou, BJ. 2:13-19. p. 122. Yaninek, J. S.; James, B. D.; Bieler, P. Jerath, M. L. 1967. A review of 50 years 1994. Ecologically Sustainable applied entomology in Nigeria: Root Cassava Plant Protection (ESCaPP): A crops. Entomol. Soc. Niger. p. 5-12. model for environmentally sound pest management in Africa. Afr. Crop Sci. J. 2:553-562.

39 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.5 Cameroon

Nelson Ntonifor*, Braima James**, Brice Gbaguidi** and Ambe Tumanteh***

Introduction Fishpool and Burban, 1994; Legg, 1995), as well as by man through the Cassava mosaic begomoviruses transport and replanting of infected transmitted by Bemisia tabaci and cassava cuttings. In Cameroon, causing cassava mosaic disease Fondong et al. (1997), using infection (CMD) are among the most important of initially CMD-free cuttings as an vector-borne pathogens of crop plants indicator, showed that the spread of in sub-Saharan Africa (Geddes, 1990). the disease was more rapid in the Yield losses attributable to CMD are lowland forest than in the mid- estimated at 28%-40% (Thresh et al., altitude forest and savannah areas. It 1994). Fauquet and Fargette (1990) was also observed that CMD spreads reported yield losses ranging from faster in monocultures of cassava or 20% to 95% for particular varieties of when cassava is intercropped with cassava (Manihot esculenta Crantz) cowpea (Vigna unguiculata [L.] Walp.) under specific conditions. The than in cassava intercropped with incidence and effects of CMD in maize (Zea mays L.) (Fondong et al., different ecozones are influenced by 1997). In a regional CMD environmental factors, the intensity of epidemiology trial covering Benin, cassava cultivation, the relative Cameroon, Ghana and Nigeria, susceptibility and sensitivity to varietal differences in the rate of CMD infection of cassava genotypes grown, infection were observed among and the virulence and abundance of improved and local varieties (James et virus and vector species (Fargette and al., 1998). Thresh, 1994). Disease epidemiology may be influenced also by the Given that many factors influence incidence of Bemisia tabaci CMD epidemiology, a detailed (Gennadius), the principal vector of understanding of the various cassava CMD, and of B. afer (Priesner and agro-ecologies, the associated whitefly Hosny), a non-vector species that also species composition and the occurs on cassava (Robertson, 1987; cultivation practices of the farmers may be helpful in designing the most appropriate CMD control strategy.

* Department of Biological Sciences, Furthermore, since the disease is University of Buea, Cameroon. prevalent in almost all the major ** International Institute of Tropical cassava production areas of Africa, Agriculture (IITA), Biological Control Center potential advantages are to be gained for Africa, Cotonou, Benin. *** Institute of Agricultural Research for in collecting such baseline data from Development, Ekona, Cameroon. different areas using a common

40 Cameroon protocol, to enable valid comparisons Increased Biological to be made between countries and thus Understanding facilitate the search for effective solutions. This chapter reports the Two whitefly species, B. tabaci and status of the CMD problem, the species B. afer, were identified on cassava in composition and relative abundance of Cameroon. B. tabaci was the prevalent Bemisia whitefly populations, and the species, comprising at least 92% of cassava production practices in three whitefly adult collections in each of the major cassava agro-ecologies of zones surveyed. Whitefly abundance Cameroon. averaged 4.1 adults per cassava shoot tip in the rainforest, 3.1 in the transition Surveys carried out in November forest and 2.4 in the wet savannah. and December 1997, and later in Logarithm-transformed whitefly counts March 1998, followed the methods were significantly lower in the wet previously agreed among project savannah than in each of the other partners. Seventy fields were surveyed ecozones (t > 3.11; P < 0.05). Two in three cassava growing regions parasitoid species, Encarsia sophia chosen to represent the major agro- (Girault and Dodd) and Eretmocerus sp. ecological regimes under which cassava were hatched from Bemisia mummies; is grown in Cameroon: South-West E. sophia predominated in the Province, representing rainforest collections. (26 fields); Central/South Province, representing transition forest Total CMD incidence averaged 44% (28 fields); and North-West Province, in the rainforest, 72% in the transition representing wet savannah (16 fields) forest and 45% in the wet savannah. (Figure 1). Incidence of stem cutting infection contributed far more to total CMD incidence than did whitefly infection (Figure 2) and was significantly higher in the transition forest than in either of the other ecozones (t = > 3.09; P < 0.05). A possible reason for the high incidence of

Nigeria cutting infection observed in the transition forest is that at least 90% of farmers in the region grow CMD- susceptible local cassava varieties whose tender leaves are consumed as leafy vegetables. The incidence of whitefly Cameroon infection, transformed to allow for the North- effect of multiple infection (Gregory, West Central 1948), was significantly lower in the South- African transition forest than in either of the West Republic other ecozones (t > 6.27; P < 0.05). In all ecozones, most of the diseased plants South- Central had a mean damage severity score of 2 (“slight damage”, on an ascending damage severity scale of 1 to 5) and the Eq, guinea Gabon Congo proportion of plants with damage severity score 4 was negligible; there Figure 1. Areas surveyed for whitefly incidence and cassava mosaic were no plants with damage score 5 disease in Cameroon. (Figure 3).

41 Whiteflies and Whitefly-borne Viruses in the Tropics

80 sites). Fourteen percent of the farms 70 were on land that had been worked for 60 between 2 and 5 years, and 26% on 50 land that had been cultivated for less 40 than 2 years. Only 21% of cassava 30 farmers practiced crop rotation. 20 Nineteen percent of farmers ranked 10 Percentage of plants with CMD symptoms cassava as the most profitable crop and 0 Rainforest Transition Wet 90% of farmers sold all or part of their forest savannah cassava crop. Whitefly infection Cutting infection Farmers generally attached no Figure 2. Cassava mosaic disease (CMD) incidence and source of infection in significance to whiteflies and in their the ecozones of Cameroon. local languages they simply referred to them as “flies” or “white insects”. In contrast, 51% of farmers recognized 80 CMD as a problem and in local 70 languages referred to the disease 60 symptoms variously as “crazy disease”, 50 “curl”, “jelly cassava”, “leaf curl”, 40 “leprosy”, kalle, khumbo, nome or 30 20 pama. Use of these local names, which 10 describe the symptoms, will be helpful different CMD severities

Percentage of plants with 0 in training farmers to improve their Slight Moderate Serious understanding of the causes and damage damage damage nature of the disease. Rainforest Transition forest Wet savannah No farmer reported total yield losses attributable to CMD; however, Figure 3. Cassava mosaic disease (CMD) 16% of farmers reported losses of 25% damage severity in the ecozones of Cameroon. in the first season, 20% reported 50% losses and 10% reported 75% losses. These figures are surprising in view of the mild damage observed in the Increased Socio-Economic survey, which, based on experience Understanding elsewhere, would not be expected to lead to significant yield loss. Most Cassava farmers were mostly women farmers indicated that climate affected (87%). Sixty-six percent of farmers whitefly/CMD problems and stated owned land whilst 20% rented land for that the problem was most severe farming; the remaining 14% offered no during the early growth stages of response on the issue of land tenure. cassava. This period corresponds to Farm size averaged 0.29 hectares March-April during the first season and (2.5 nearby fields) in South West to August during the second growing Province, 0.6 hectares (2.4 nearby season in Cameroon. Even though fields) in Central/South Province and most farmers recognized CMD as a 0.32 hectares (4.2 nearby fields) in problem that occurs every year, 61% of North-West Province. Farmers grew them had received no technical cassava mostly on land that had been information or assistance on the worked for more than 5 years (60% of problem.

42 Cameroon

Farmers undertook various CMD-resistant varieties. Farmer-led measures to combat the whitefly/CMD rapid multiplication schemes to problem. The common practices were produce CMD-resistant varieties may roguing, choice of resistant varieties, therefore be relatively easy to promote selection of planting material and the under such circumstances by building application of wood ash or of fertilizers upon previous technology transfer (Figure 4). Less than 2% of farmers experiences and farmer training used pesticides against whiteflies and/ programmes in the country (Dahniya or CMD. Even though it might be et al., 1994; Akoroda, 1997; Bakia et supposed that the high incidence of al., 1999). In selecting planting cutting infection observed in the survey material, the five most important would make roguing an unattractive criteria were yield (34.3% of farmers), strategy, it was nevertheless the followed by stem health (25.7% of predominant practice. farmers), absence of diseases (17.1% of farmers), stem size (5.7%) and stem maturity (4.3%). Most farmers Roguing expressed willingness to change Varieties planting dates for CMD control but Selection 30% of them indicated unwillingness Fertilizers to alter planting date. Wood ash Pesticides 0 5 10 15 20 25 30 35 Conclusions Relative frequency (%) of practice There was only limited evidence of Figure 4. Cassava mosaic disease management practices used by farmers in CMD being spread by its whitefly Cameroon. vector in farmers’ fields in Cameroon. Infected planting material was the major source of CMD and a major In their choice of varieties, 86% of extension effort will be required to the farmers were aware of the encourage farmers to select disease- differences in susceptibility to the free and healthy planting materials. If disease among cassava varieties. effective measures to improve crop Among improved varieties health are to be developed, the recommended for their disease importance of latent infection will have resistance, “Agric” (of International to be investigated. The expressed Institute of Tropical Agriculture, Nigeria willingness of farmers to alter planting [IITA] origin), chosen by 47% of dates and their conviction that CMD farmers, was the most commonly incidence varies with climatic selected. The variety Pawpaw leaf was conditions paves the way for the single most frequently planted participatory action research to CMD-resistant local variety (chosen by optimize times of planting in the 9% of farmers). Other CMD-resistant various ecozones so as to reduce CMD local varieties cited were Metta agric, incidence and damage. In such Mfont and Ndongo. studies, it may be useful to monitor the population dynamics of B. tabaci to Most farmers (96%) indicated that see whether its abundance varies at in selecting planting materials they the different planting times. Evaluation intentionally chose parent plants with of parasitism by existing natural minimal or without CMD symptoms. enemies (E. sophia, for example) may This practice would inherently favour help show whether biological control

43 Whiteflies and Whitefly-borne Viruses in the Tropics contributes significantly to regulating Fargette, D.; Thresh, J. M. 1994. The populations of the vector. Such ecology of African cassava mosaic information will contribute to geminivirus. In: Blakeman, J.P.; Williamson, B. (eds.). Ecology of understanding the role of B. tabaci in plant pathogens. CAB International, the epidemiology of CMD. Wallingford, GB. p. 269-282.

Even though the CMD damage Fauquet, C.; Fargette, D. 1990. African observed in the survey was slight, cassava mosaic virus: Etiology, farmers provided high yield loss epidemiology and control. Plant Dis. estimates. This suggests the need for 74:404-411. farmer participatory yield loss trials to Fishpool, L. D. C.; Burban, C. 1994. quantify losses both of storage roots Bemisia tabaci: The whitefly vector of and leaves, especially since the latter African cassava mosaic geminivirus. are widely consumed as leafy Trop. Sci. 34:55-72. vegetables, but losses have been little studied. Farmer participatory varietal Fondong, V. N.; Adipala, A.; Thresh, J. M. selection trials also need to be 1997. Spread of African cassava geminivirus at four agroecological conducted on the local cassava locations of Cameroon. Afr. Crop varieties (such as Pawpaw leaf ), which Sci. J. 3:1035-1049. farmers report to be CMD-resistant, and comparisons made with cultivars Geddes, A. M. 1990. The relative whose response to CMD is well importance of crop pests in sub- documented. Such studies will provide Saharan Africa. Bulletin no. 36, valuable input into breeding Natural Resources Institute (NRI), Chatham, GB. programmes that aim to achieve sustainable high yields in varieties Gregory, P. H. 1948. The multiple preferred by farmers and to increase infection transformation. Ann. Appl. adoption of such varieties. Biol. 35:412-417.

James, B. D.; Legg, J. P.; Gbaguidi, J. B.; Annang, D.; Asiabaka, C. C.; References Cudjoe, A. R.; Echendu, T. N. C.; Maroya, N. G.; Ntonifor, N.; Ogbe, Akoroda, M. O. 1997. Cassava technology F.; Salawu, R. A.; Tumanteh J. A. transfer: Lessons from Cameroon 1988. Understanding farm-level and Nigeria. Afr. J. Root Tuber epidemiology of cassava mosaic Crops 2:257-260. disease. In: Abstracts, 7th Triennial Symposium of the International Bakia, B.; Ambe, J. T.; James, B. 1999. Society for Tropical Root Crops- Technology transfer strategies: The Africa Branch (ISTRC-AB), 11-17 case of sustainable cassava plant October 1988, Cotonou, BJ. p. 122. protection in Cameroon. Afr. J. Root Tuber Crops 3(2):23-27. Legg, J. P. 1995. The ecology of Bemisia tabaci (Gennadius) (Homoptera: Dahniya, M. T.; Akoroda, M. O.; Alvarez, Aleyrodidae), vector of African M. N.; Kaindaneh, P. M.; Ambe- cassava mosaic geminivirus in Tumanteh, J.; Okeke, J. E.; Jalloh, Uganda. Ph.D. thesis, University of A. 1994. Development and Reading, GB. dissemination of appropriate root crops packages to farmers in Africa. In: Ofori, F.; Hahn, S. K. (eds.). Tropical root crops in a developing economy. Procs. Ninth Symposium of the International Society for Tropical Root Crops (ISTRC), 20-26 October 1991, Accra, GH. p. 2-9. 44 Cameroon

Robertson, I. A. D. 1987. The role of Thresh, J. M.; Fishpool, L. D. C.; Otim- Bemisia tabaci (Gennadius) in the Nape, G. W.; Fargette, D. 1994. epidemiology of ACMV in East African cassava mosaic virus Africa: Biology, population dynamics disease: An under-estimated and and interaction with cassava unsolved problem. Trop. Sci. varieties. In: African cassava mosaic 34:3-14. disease and its control, Procs. International Seminar, 4-8 May 1987, Yamoussoukro, CI. Centre Technique de Coopération Agricole et Rurale, Wageningen, NL, and Institut Français de Recherche Scientifique pour le Développement en Coopération, Paris, FR. p. 57-63.

45 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.6 Uganda

Peter Sseruwagi*, James Legg** and William Otim-Nape***

Introduction with the CMD epidemic, coupled with drought, led to food shortages and Cassava mosaic disease (CMD) is the localized famine (Otim-Nape et al., most important viral disease of cassava 1997). Between 1994 and 1998, the (Manihot esculenta Crantz) (Otim-Nape, CMD epidemic continued to expand, 1993) in Uganda. It is caused by affecting much of southern Uganda. begomoviruses transmitted by the Significant progress has been made in whitefly vector, Bemisia tabaci controlling the disease, particularly in (Gennadius) and spread through north-eastern parts of the country, virus-infected planting material which were some of the first to be (Harrison, 1987). affected (Thresh et al., 1994) but the epidemic continues to spread in The first recording of CMD in southern and western districts. The Uganda was in 1928 (Martin, 1928). principal control method has been the Severe epidemics from 1933 to 1944 deployment of resistant varieties. were successfully controlled by planting resistant cassava varieties and Sweetpotato virus disease (SPVD) is through phytosanitation measures correspondingly the most serious viral enforced by local government statute disease of sweetpotato (Ipomoea batatas (Jameson, 1964). The disease then was [L.] Lam.) in Uganda (Bashaasha et al., considered a relatively minor problem 1995), although its distribution and until the late 1980s, when severe impact on sweetpotato production have epidemics were again reported in scarcely been studied until recently. northern Luwero District (Otim-Nape, Aritua et al. (1998) provide information 1988). By 1990, CMD had severely on the disease for only two of the affected other cassava-growing districts country’s (at that time) 45 districts, of northern and eastern Uganda (Otim- Mpigi and Soroti (Figure 1). SPVD Nape et al., 1997). In 1993 and 1994, results from co-infection of Sweetpotato the widespread crop losses associated chlorotic stunt virus, transmitted by B. tabaci and Sweetpotato feathery * International Institute of Tropical mottle virus, transmitted by the aphid Agriculture-Eastern and Southern Africa Myzus persicae (Sulzer) (Gibson et al., Regional Center (IITA-ESARC), Kampala, 1998). Uganda. ** IITA-ESARC, Kampala, Uganda, and Natural Resources Institute, University of In order to enhance understanding Greenwich, Chatham Maritime, Kent, UK. of whiteflies and whitefly-transmitted *** Namulonge Agricultural and Animal Production Research Institute (NAARI), viruses on cassava and sweetpotato in Kampala, Uganda. Uganda, surveys were conducted

46 Uganda between November 1997 and January Increased Biological 1998. It was anticipated that the data Understanding collected would be useful in helping researchers in Uganda monitor the development of the CMD epidemic, Whitefly species and abundance target control interventions more The survey identified two whitefly effectively and, in the longer term, species, B. tabaci and B. afer (Priesner develop sound integrated pest and Hosny). Of the 261 samples management (IPM) strategies for identified on cassava, 95.8% were cassava and sweetpotato. B. tabaci and 4.2% B. afer. Only B. tabaci was recorded from sweetpotato.

Whitefly populations on cassava (counted on the top five leaves for a Uganda single shoot of each of 30 plants per sampled field) were highest in the south- Lira western districts of Masaka and Rakai Apac Masindi and lowest in Masindi (the north- western limit of the survey area) and Rallisa LuweroMukono Tororo (on Uganda’s eastern border with Tororo Mubende Iganga Kenya), with mean abundance ranging Mpigi from 0.7 to 13.1 per top five leaves Masaka (Table 1) (see also Figure 2 in Chapter Lake Rakai Victoria 1.14, this volume). The pattern of whitefly populations on sweetpotato (counted by disturbing leaves and

Figure 1. Cassava- and sweetpotato-growing making 10 serial counts of 1 min each of areas surveyed for whitefly incidence whitefly adults observed) was similar to in Uganda. that on cassava: abundance was greatest in the central districts of Surveys were carried out in Mukono and Mpigi and lowest in 12 districts of Uganda (Figure 1) by Masindi and Apac. Mean adult whitefly multi-disciplinary teams from the abundance ranged from 0.9 to 14.8 per National Agricultural Research minute count. In similar studies, Aritua Organisation and the International et al. (1998) found greater whitefly Institute of Tropical Agriculture (IITA). abundance on sweetpotato in the Districts were chosen to represent southern and central zones, which are major cassava and sweetpotato characterized by more evenly distributed producing areas. Farmers’ fields of rainfall than the northern and eastern cassava 3-5 months after planting and zones. of sweetpotato 3 months after planting were selected at regular intervals along The data indicate a higher whitefly main roads traversing each district. population on both crops in southern Data were collected according to the and south-western Uganda than in the standardized survey methods, except northern and eastern districts of the that the number of fields varied country. Assessing the significance of between the districts because the these differences is difficult based on survey areas were selected on the records collected on a single occasion, basis of agro-ecological differences during what appears to be the rather than administrative boundaries continuing spread of an epidemic. (Table 1). Numerous factors such as physical

47 Whiteflies and Whitefly-borne Viruses in the Tropics a ance on cassava and Sweetpotato ) and on sweetpotato (per minute gory, P. H. 1948. The multiple counts incidence (1-5 scale) (1-5 scale) a (47)(59)(29) 64 61 7 77 78 31 3.0 2.8 2.7 14.8 14.7 3.8 10 29 9 2.4 2.7 2.9 11 (19)11 16 (73) 34(28)21 (56)12 45 69 15 3.0 85 71 36 3.6 83 0.9 3.0 2.6 3 1.5 3.4 2.0 7.8 15 6 2.8 7 2.6 2.2 Whitefly Cutting Total Severity Incidence Severity infection infection Ann. Appl. Biol. 35:412-417); severity of disease is measured on an ascending 1-5 scale, from low to severe. counts Whitefly CMD Whitefly SPVD . fields sweetpotato in selected districts of Uganda, surveyed during November and December 1997. count); whitefly infection, figures in parentheses transformed to multiple infection units allow for (Gre infection transformation District No. Cassava MasindiApacLiraIganga 7TororoPallisa 6Mukono 0.7 7 6Mpigi 7Luwero 1.4 7Mubende 7 1.4 1.6Masaka 0.9 7Rakai 6 3.0 7(140) 22 3.7 Average 6(89) 11 (80) 20 4.6 3.0 3.4 7 71 13 13.1 81 3.9 11.1 17 64 24 93 50 (91) 92 84(65) 22 (60) 44 3.0 17 3.0 2.8 3 46 1.9 67 11.2 1.4 68 47 9 3.0 16 3 2.9 2.7 2.3 3.2 2.1 2.3 5.5 2.0 14 8 11 2.7 2.5 2.7 Table 1. Incidence (%) of cassava mosaic disease (CMD) and sweetpotato virus (SPVD), severity whitefly abund a. Figures are means for each district. Whitefly counts, whitefly abundance on cassava (number of whiteflies per top five leaves

48 Uganda environment, host plant health in southern areas along the shoreline of condition and natural enemies have Lake Victoria, with the highest incidence been shown to influence whitefly in Mpigi District on the northern shore populations on cassava (Legg, 1995). of the lake, and the lowest in Masindi Nonetheless, this general geographical and Lira, further to the north. Overall, trend in abundance does parallel the SPVD symptom severity was moderate current trend in disease severity. in all the districts surveyed.

Disease incidence and Natural enemy species and symptom severity distribution CMD was present in all cassava fields Whitefly natural enemies collected in the sampled and overall incidence (the survey included predators and average proportion of affected plants) parasitoids. Few predators were throughout the survey was 68% (see collected. This was probably due to the also Figure 2 in Chapter 1.14, this limited time available during the survey volume). Disease incidence was highest for collecting samples, the relatively low in the northern districts of Apac and populations of whiteflies on cassava and Lira and lowest in Mubende. Symptoms sweetpotato, and the difficulty of were moderate (2.6 to 3.0 on an observing arthropods actually feeding on ascending scale of 1 to 5) with the whitefly immature or adult stages. exception of Tororo District, where the disease was severe (Table 1). The predator Conwentzia africana Meinander was recorded from 16 locations in three of the four target The pattern of CMD incidence in areas. Three species of whitefly Uganda indicates a north-to-south parasitoids were identified: Encarsia progression. In areas of epidemic sophia (Girault and Dodd), Encarsia sp. expansion in the south (Masaka and (luteola group) and Eretmocerus sp. Rakai Districts) the main source of E. sophia was the most widespread and infection is whitefly transmission, numerous, occurring through most of whereas in post-epidemic areas to the the surveyed area (Table 2). Eretmocerus north (Masindi, Lira and Apac Districts) sp. occurred principally in south- whitefly transmission is less evident western Uganda. Encarsia sp. (luteola and diseased cuttings are the main group) was only identified from eastern source of infection (Table 1). The Uganda. The absence of records of relatively low CMD incidence recorded parasitoids from some of the districts in Pallisa District is because of the surveyed, particularly those in the National Cassava Programme’s recent north, is thought to be largely because introduction of CMD-resistant varieties, of the relatively low abundance of Nase 1 (TMS 60142), Nase 2 whiteflies recorded in these locations. (TMS 30337) and Nase 3 (TMS 30572), Seasonal and environmental factors originating from the IITA breeding clearly play an important role in the programme. Similarly, disease abundance of whiteflies and their incidence and infection were relatively parasitoids, and in order to collect more low for the districts neighbouring comprehensive data, the surveys would Pallisa-Kumi, immediately to the north, have to be repeated, perhaps at and Soroti, to the south. 3-monthly intervals throughout the year. Little is known about the role of SPVD occurred in 60 of the natural enemies in the population 80 sweetpotato fields sampled. dynamics of B. tabaci on cassava and Generally, SPVD incidence was greater sweetpotato.

49 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 2. Occurrence of parasitoids reared SPVD, with 65% of the producers who from mummies collected from recognized SPVD as a problem cassava at each of the sampling sites in Uganda. considering that it was becoming more District Parasitoid occurrence per severe. Farmers expressed no clear sampling sitea opinion as to the time of year at which Encarsia Eretmocerus Encarsia sp. whitefly/disease problems are sophia sp. (luteola group) especially severe. The picture of the Masindi - - - spread of the CMD epidemic provided Apac - - - by farmers’ first record of severe CMD Lira - - - in their fields, indicating a north-to- Iganga 3/6 1/6 1/6 south progression, is in close Tororo 2/7 - - agreement with scientific records of the Pallisa 2/7 - - expansion of the severe CMD epidemic Mukono 6/7 1/7 - (Otim-Nape et al., 1997; Legg and Mpigi 4/7 3/7 - Ogwal, 1998). Luwero - - - Mubende 2/7 - - Managing whiteflies and Masaka 1/6 3/6 - whitefly-transmitted viruses Rakai 3/7 2/7 - A common response amongst farmers a. Dashes indicate the absence of any reared to the CMD problem has been to parasitoids. abandon production of the crop altogether. Whilst cassava farmers assessed losses to CMD at more than Increased Socio-Economic 75% of production, sweetpotato Understanding farmers estimated losses to SPVD to be less than 25%. Farmers attempting to control CMD and SPVD mainly used Farmers’ assessment of roguing, while some applied wood ash whitefly-related problems on the infected plants. Only 6.5% of Only a small proportion of cassava and farmers reported having received sweetpotato farmers (22%) recognized information on control practices whiteflies. In contrast, the diseases through extension and/or research CMD and SPVD were more widely channels. Selection of planting material recognized (CMD by 100% of cassava free of disease symptoms was the most farmers and SPVD by 48% of widely used disease control practice for sweetpotato farmers) and had specific both cassava and sweetpotato, and the local names in most of the locations second most widely reported was where they were prevalent. Both roguing of plants that developed diseases were commonly referred to symptoms. However, most farmers with words describing the mottling or reported the two methods to be only deformation of the leaves induced by partially effective in controlling the the disease, and a number of names diseases. were common for both crops. Virtually all cassava farmers recognized CMD as Thirteen out of 19 farmers growing a production constraint, whereas for resistant varieties of cassava and zero SPVD only 35% of producers out of 11 farmers growing resistant considered it a problem. Most cassava sweetpotato varieties cited resistance farmers (74%) considered that CMD as a control method. However, only was becoming more severe. about 24% of cassava farmers growing Sweetpotato producers were similarly resistant varieties found their pessimistic about the outlook for performance to be superior to that of

50 Uganda local varieties. Assessments of the provide an understanding of the benefits of disease-resistant economic costs and benefits of sweetpotato varieties were more mixed, managing these diseases. with some farmers reporting no difference in performance between resistant and local varieties. About 20% of both cassava and sweetpotato Conclusions farmers attributed differences in the The results of the survey indicate that disease response of the different farmers’ approaches to managing the varieties to some kind of resistance to whitefly-borne virus diseases, CMD and infection, inherent in the variety. On SPVD, are generally ad hoc and based the other hand, 28% of cassava on only a partial understanding of the producers and 18% of sweetpotato problem. Since researchers already producers had no idea why such have a considerable body of knowledge differences should occur. Very few relating to these problems that is not farmers reported using chemical being applied at farm level, this pesticides to control either CMD or suggests that information flow between SPVD. Of these farmers, almost all farmers and researchers is weak. used dimethoate, an organophosphate However, the absence of a clearly insecticide, so their intention was defined and well-articulated IPM presumably to control the whitefly approach to managing CMD and SPVD vector. Most of these farmers used could be a contributory factor. pesticides on the recommendation of Researchers believe that such an neighbours or relatives, sprayed when approach would combine the use of symptoms were seen, and did so on both local and improved resistant one to three occasions during the varieties with the selection of clean production cycle. However, many planting material and subsequent reported that the approach was roguing of diseased plants—plant ineffective. health methods that are already widely practiced. Whitefly-related crop production costs in affected The development and areas enhancement of channels of Since farming families produce most information flow from research to farm cassava and sweetpotato for home level is clearly critical to the successful consumption rather than sale, it was adoption of IPM of whiteflies and difficult to assess the value of lost whitefly-transmitted viruses. A number production attributable to CMD and of initiatives have recently been SPVD and costs incurred in trying to implemented in Uganda to strengthen manage these diseases. As mentioned both the extension system and linkages above, the most widely practiced between research and extension and to management tactics were selection of foster a participatory approach to disease-free planting material and working with farmers to develop roguing, both of which involve appropriate and sustainable IPM investment of additional labour. strategies. It is hoped that these will Because labour is usually provided by improve the efficiency of the two-way members of farming households, rather flow of information between researchers than hired, this represents a time- and farmers in the years ahead, related opportunity cost rather than a facilitating the more widespread direct financial one. More focused adoption of IPM strategies as they follow-up studies are required to become available.

51 Whiteflies and Whitefly-borne Viruses in the Tropics

Future work should focus mainly also acknowledge Dr. Mohammed Ali on managing CMD and SPVD, through Bob of the International Center of multiplication and distribution of the Insect Physiology and Ecology for his existing disease-resistant varieties, and identification of whitefly species and on developing, evaluating, multiplying natural enemies. and distributing new disease-resistant materials. Since most farmers reported the planting of clean cuttings and References roguing of diseased plants to be ineffective, there is a need to find out Aritua, V.; Adipala, E.; Carey, E. E.; under what conditions these measures Gibson, R. W. 1998. The incidence of sweetpotato virus disease and could be both effective and adoptable virus resistance of sweetpotato by farmers. Future work should also grown in Uganda. Ann. Appl. Biol. consider the possibility of community- 132:399-411. based phytosanitation, which offers the prospect of widespread reductions in Bashaasha, B.; Mwanga, R .O. M.; Ocitti CMD or SPVD inoculum pressure and p’Obwoya, C. O.; Ewell, P. T. 1995. thereby reduced rates of spread. Too Sweetpotato in the farming and food systems of Uganda: A farm survey little is currently known about the report. Centro Internacional de la activity of parasitoids or other natural Papa (CIP), sub-Saharan Africa enemies of B. tabaci on either cassava Region, Nairobi, KE/National or sweetpotato to offer the prospect of Agricultural Research Organisation their immediate use within an IPM (NARO), Kampala, UG. 63 p. strategy. However, both introductions of exotic parasitoids and augmentation Gibson, R. W.; Mpembe, I.; Alicai, T.; Carey, E. E.; Mwanga, R. O. M.; through habitat management have Seal, S. E.; Vetten, H. J. 1998. been used successfully in the United Symptoms, aetiology and serological States to improve management of analysis of sweetpotato virus B. tabaci (Goolsby et al., 2000). It may disease in Uganda. Plant Pathol. be appropriate therefore to explore 47:95-102. similar approaches to using parasitoids such as E. sophia for the management Goolsby, J. A.; Ciomperlik, M. A.; Kirk, A. A.; Jones, W. A.; Legaspi, B. C.; B tabaci of . as vector of CMD and Legaspi, J. C.; Ruiz, R. A.; Vacek, D. SPVD, although this is always likely to C.; Wendel, L. E. 2000. Predictive be a measure of secondary importance and empirical evaluation for to the major control approaches of parasitoids of Bemisia tabaci resistant varieties and phytosanitation. (Biotype “B”), based on morphological and molecular systematics. In: Austin, A.; Dowton, M. (eds.). Hymenoptera: Evolution, Acknowledgements biodiversity, and biological control. 4th International Hymenopterists The technical support of Dr. Nicole Conference, 1999, Canberra. Smit of the Centro Internacional de la Commonwealth Scientific and Papa (CIP), Dr. Benson Odongo, Industrial Research Organisation Mr. Valentine Aritua, Mr. Titus Alicai, (CSIRO), Collingwood, Victoria, AU. Mr. William Sserubombwe and p. 347-358. Mr. Solomon Ogwal of the National Gregory, P. H. 1948. The multiple Agricultural Research Organisation of infection transformation. Ann. Appl. Uganda and Mr. Geoffrey Okao-Okuja Biol. 35:412-417. of IITA, is acknowledged. The authors

52 Uganda

Harrison, B. D. 1987. Properties and Martin, E. F. 1928. Report of the geographical variation of mycologist. In: Annual Report of the geminivirus isolates from mosaic- Department of Agriculture. affected cassava. In: African cassava Government Printer, Entebbe, UG. mosaic disease and its control. p. 31. Procs. International Seminar, 4-8 May 1987, Yamoussoukro, CI. Otim-Nape, G. W. 1988. Cassava Centre Technique de Coopération situation in Luwero district. A Agricole et Rurale, Wageningen, NL, mission report by the Uganda Root and Institut Français de Recherche Crops Programme. Namulonge Scientifique pour le Développement Research Station, Kampala, UG. en Coopération, Paris, FR. p. 270. Otim-Nape, G. W. 1993. Epidemiology of Jameson, J. D. 1964. Cassava mosaic the African cassava mosaic disease in Uganda. East Afr. Agric. geminivirus disease (ACMD) in For. J. 29:208-213. Uganda. Ph.D. thesis, University of Reading, GB. Legg, J. P. 1995. The ecology of Bemisia tabaci (Gennadius) (Homoptera: Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; Aleyrodidae), vector of African Baguma, Y.; Ogwal, S.; Semakula, cassava mosaic geminivirus in G. N.; Acola, G.; Byabakama, B.; Uganda. Ph.D. thesis, University of Martin, A. 1997. Cassava mosaic Reading, GB. virus disease in Uganda: The current pandemic and approaches Legg, J. P.; Ogwal, S. 1998. Changes in to control. Natural Resources the incidence of African cassava Institute (NRI), Chatham, GB. 65 p. mosaic geminivirus and the abundance of its whitefly vector Thresh, J. M.; Fishpool, L. D. C.; Otim- along south-north transects in Nape, G. W.; Fargette, D. 1994. Uganda. J. Appl. Entomol. 122: African cassava mosaic virus 169-178. disease: An under-estimated and unsolved problem. Trop. Sci. 34:3-14.

53 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.7 Kenya

Joseph Kamau*, Peter Sseruwagi** and Valente Aritua***

Introduction (1958) and resulted in the release of CMD-resistant hybrids to research Cassava mosaic disease (CMD) is stations in the country. However, caused by cassava mosaic farmers did not replace their local begomoviruses (CMBs), which are traditional Kenyan cultivars of cassava transmitted by the whitefly Bemisia (Manihot esculenta Crantz) with this tabaci (Gennadius). Further spread may improved germplasm, except for occur through farmers distributing and 46106/27 “Kaleso”, which became planting virus-infected cuttings widely grown in Coast Province (Bock, (Harrison, 1987). The disease causes 1994b). higher production losses than any other virus disease of cassava in Kenya (Bock, Research on CMD and its whitefly 1994b). Sweetpotato virus disease vector only became important in Kenya (SPVD) results from co-infection of during the 1970s, when a major plant sweetpotato (Ipomoea batatas [L.] Lam.) virology project was initiated. The work by two distinct viruses: Sweetpotato of this project initially emphasized the chlorotic stunt virus, transmitted by B. isolation of the putative Cassava mosaic tabaci, and Sweetpotato feathery mottle virus (Bock, 1975), which was later virus, transmitted by the aphid, Myzus shown to cause CMD (Bock and Woods, persicae (Sulzer) (Gibson et al., 1998). 1983). Subsequent epidemiological Although SPVD is the most prevalent work by Bock (1983) noted that the viral disease of sweetpotato in farmers’ incidence of CMD was generally high in fields in Kenya (Carey et al., 1998), its coastal and western Kenya, where it effects on growth and yield have yet to exceeded 80% in some districts and be quantified. approached 100% on individual farms but the spread of disease into initially The early history of CMD in Kenya CMD-free plantings was generally slow. is obscure but work on cassava under Additional studies associated with this the East African breeding programme in project were the description of the the mid-1950s is described by Doughty population dynamics of B. tabaci (Seif, 1982; Robertson, 1987), assessments of yield loss (Seif, 1981) and the * Kenya Agricultural Research Institute (KARI), National Dryland Farm Research development of control methods (Bock, Center, Katumani, Kenya. 1994a). During the 1990s, a series of ** International Institute of Tropical surveys conducted to reassess the Agriculture-Eastern and Southern Africa Regional Center (IITA-ESARC), Kampala, incidence of the disease in the country Uganda. indicated increased spread and *** Makerere University, Kampala, Uganda. incidence of CMD in farmers’ fields.

54 Kenya

Recent epidemiological studies on and to assess producers’ knowledge of CMD in the region have provided whiteflies and the diseases they evidence of the spread of severe CMD transmit on cassava and sweetpotato. from Uganda to western Kenya (Legg, Three “target areas” representing the 1999). This is associated with the major cassava and sweetpotato growing spread of a novel and particularly areas in the country were selected: virulent CMB (Harrison et al., 1997; Coast Province, Western Province and Legg, 1999). Major losses have been Nyanza Province (Figure 1). Within experienced already in Western these areas, 3 to 5-month-old cassava Province, and districts of Nyanza fields and 3-month-old sweetpotato Province are now threatened. Efforts fields were randomly selected at have been made to control the disease regular intervals along main roads. by introducing resistant material into western Kenya from the breeding program of the International Institute Sudan Ethiopia of Tropical Agriculture-Eastern and Southern Africa Regional Center (IIITA- ESARC) in Namulonge, Uganda. Continuing work by the cassava Uganda program of the Kenya Agricultural Kenya Research Institute (KARI) emphasizes the evaluation and multiplication of the Western Somalia initial stock of resistant material, in collaboration with international Nyanza research institutions and networks such as the East Africa Root Crops Coast Research Network (EARRNET) (Legg et al., 1999). Tanzania

Research on SPVD began during the 1950s, when Sheffield (1957) reported the presence of a severe Figure 1. Cassava- and sweetpotato-growing sweetpotato mosaic disease on areas surveyed for whitefly incidence in Kenya. sweetpotato fields in East Africa. More recently, surveys have been carried out to assess the occurrence and importance of sweetpotato viruses in Increased Biological Kenya (Carey et al., 1998) and the Understanding Centro Internacional de la Papa (CIP) now has a major germplasm Whitefly species and introduction and development program abundance based in Kenya. Whilst cultivar resistance has been identified as the Whitefly nymph samples were obtained principal strategy for managing SPVD, from most but not all surveyed sites little detailed research has been done and, for each sample, from one to three on yield loss or the epidemiology of the nymphs were identified to species level. disease. Two whitefly species, B. tabaci and Bemisia afer (Priesner and Hosny), were A study was conducted in mid- identified on both cassava and 1998 to identify whiteflies and whitefly- sweetpotato. Seventy whitefly nymph transmitted viruses prevalent in Kenya samples were identified from cassava;

55 Whiteflies and Whitefly-borne Viruses in the Tropics of which 58 were B. tabaci and 12 were farmers’ fields. The disease was more B. afer. Only 35 whitefly samples were prevalent in Western and Nyanza identified from sweetpotato Provinces than in Coast Province. (attributable to the lower populations on that crop), comprising 31 B. tabaci Whitefly parasitoids and four B. afer. The two species The survey identified two species of occurred throughout the three target aphelenid parasitoids; 15 individual areas. parasitoids were identified, comprising 13 Encarsia sophia (Girault and Dodd) Adult whitefly populations were and two Eretmocerus sp. Little is relatively low on both cassava and known, however, about the role of sweetpotato (Table 1). Whiteflies on natural enemies in the population cassava were more numerous in Coast dynamics of whiteflies in Kenya. Province, while whiteflies on sweetpotato were more abundant in Nyanza Province. Mean adult whitefly Increased Socio-Economic numbers ranged from 0.2 to 2.9 per Understanding top five leaves on cassava and 0.5 to 4.4 per minute count on sweetpotato in the surveyed regions (see also Figure 2, Farmers’ assessment of Chapter 1.14, this volume). whitefly-related problems About half of the producers surveyed Disease incidence and (52% of cassava farmers and 46% of symptom severity sweetpotato farmers) were able to Mean CMD incidence in the surveyed recognize whiteflies, although the regions was 51.3% and incidence was names they used were non-specific. Of highest in Western Province and lowest the farmers who recognized whiteflies in Nyanza (Table 1). Diseased cuttings on either crop, 61% considered them a provided the main source of infection production problem. in all three target areas. CMD symptom severity was mild in the Coast and Most cassava farmers (88%) Nyanza Provinces and more severe in recognized CMD as a disease of their Western Province. SPVD incidence crop, whilst only 58% of sweetpotato ranged between 3% and 53% in farmers recognized SPVD. The

Table 1. Incidence (%) of cassava mosaic disease (CMD) and sweetpotato virus disease (SPVD), disease severity and whitefly abundance on cassava and sweetpotato in three provinces of Kenya, surveyed during May and June 1998. Province Cassavaa Sweetpotatoa No. Whitefly CMD No. Whitefly SPVD fields counts fields counts incidence Whitefly Cutting Total Severity infection infection incidence Coast 17 2.9 19 (35.9) 37 56 2.3 15 0.5 0.4 Western 15 0.2 14 (65.9) 71 85 2.9 18 2.5 6.9 Nyanza 18 0.6 2 (2.3) 11 13 2.2 17 4.4 6.7 Average - 1.2 11.7 (21.6) 39.7 51.3 2.5 - 2.5 4.7 a. Figures are means for each province. Whitefly counts, whitefly abundance on cassava (number of whiteflies per top five leaves) and on sweetpotato (per minute count); whitefly infection, figures in parentheses transformed to multiple infection units to allow for multiple infection (Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. Biol. 35:412-417); severity of disease measured on an ascending 1-5 scale, from low to severe.

56 Kenya respective diseases were recognized as diseased plants and use of insecticides. a production constraint by 68% of In extreme cases, heavily infested cassava farmers and 44% of crops were simply abandoned. The sweetpotato farmers. Most cassava most widely used practices, however, farmers (82%) considered CMD to were roguing and selection. Just over occur every year, whilst only 52% of one third (38%) of cassava farmers sweetpotato farmers believed this to be used roguing, with 28% of farmers the case. Farmers had different views considering that this practice would regarding the trend of symptom reduce CMD incidence, 6% that it severity for CMD and SPVD. Fifty-four would prevent the spread of CMD and percent of cassava farmers considered 2% that it would reduce the number of that CMD severity was increasing, poorly growing plants. Among whilst only 40% of sweetpotato farmers sweetpotato farmers, 30% used considered that SPVD was becoming roguing, 18% believing that it would more important. In both cases, a reduce SPVD incidence and 8% majority of cassava (72%) and believing that it would prevent spread sweetpotato (62%) farmers believed of the disease. Both cassava and that climate influences disease severity. sweetpotato farmers rogued infected The two diseases were thought to be plants before the crop was 4 months more severe during periods of low old. Selection of clean planting rainfall by 70% of cassava farmers and material, based on the absence of 44% of sweetpotato farmers. This was disease symptoms, was used as a consistent with farmers’ views that the disease management tactic by 32% of two diseases were more severe during cassava farmers and 12% of periods of moderate temperature. sweetpotato farmers. However, even those farmers who used roguing and Yield loss estimates due to CMD selection of clean cuttings considered varied between regions. In Western these measures to be only partially Province, 21.4% of farmers estimated effective in controlling the diseases. losses at >75%. This is understandable in view of the impact that the CMD Only one cassava farmer and none pandemic has had in this area (Legg, of the sweetpotato farmers mentioned 1999). Yield loss estimates were rated chemical control as a control method. at between 25% and 50% by 79% of The cassava farmer used the farmers in Coast Province and at only organophosphate insecticide 25% by 47% of farmers in Nyanza dimethoate, applied it more than Province. At the time of the survey, the 10 times and sprayed when whiteflies CMD pandemic had affected only and damage were observed in the field. Western Province and losses recorded Few farmers reported abandoning the elsewhere were consequently less. Only production of cassava (12%) or 14% of sweetpotato farmers considered sweetpotato (2%) but many reported SPVD to be a significant source of crop occasional shortage of clean planting loss, estimating these losses at only material. None of the farmers grew 25%. improved varieties, although 42% of cassava farmers and 20% of Managing whiteflies and sweetpotato farmers noted differences whitefly-transmitted viruses in response to CMD and SPVD between Cassava and sweetpotato producers varieties. However, most farmers who reported a number of management noted differences between varieties had tactics, including selection of disease- no idea what caused the differences. free planting material, roguing of Only 4% of cassava farmers and 2% of

57 Whiteflies and Whitefly-borne Viruses in the Tropics sweetpotato farmers considered the already in neighbouring Uganda (Otim- differences to be the result of Nape et al., 1997) with the focus on the differences in levels of resistance to deployment of host plant resistance disease. Very few cassava (10%) and supported by some phytosanitation. In sweetpotato (4%) farmers had received the immediate future, some of this any technical assistance in the experience clearly needs to be management of whiteflies and whitefly- transferred to the situation in western transmitted viruses on their crops. Kenya where the CMD pandemic However, more than 30% of farmers of continues to expand. CMD-resistant each crop were willing to monitor germplasm will need to be introduced whitefly and disease problems if it from Uganda to Kenya and evaluated in would help in the management of participatory trials with farmers, whilst either CMD or SPVD. at the same time encouraging the maintenance of the quality of planting material through the use of selection Conclusions and roguing in multiplication plots. Given the evident shortcomings in the Farmers noted that CMD and SPVD understanding of both CMD and SPVD are becoming increasingly important in Kenya, participatory learning should constraints to cassava and sweetpotato be a key feature of any future control production in the country. Some programs. SPVD is clearly not as acute farmers were attempting to control the a problem in Kenya as is CMD. diseases, mainly using roguing and However, the report by farmers that selection, initiatives that in general SPVD is becoming more important is a they considered to be developed from concern and future work should aim their own knowledge. However, farmers therefore at identifying resistant recognized the methods employed in germplasm and evaluating this and controlling the two diseases as being possible phytosanitary approaches to only partially effective. During this SPVD control through participatory survey in 1998, no farmer reported research with farmers. using resistant varieties. The limited understanding that farmers demonstrated of the causes and effects Acknowledgements of CMD and SPVD, coupled with their reliance on their “own” control The technical support of Mr. Nzioki and methods, suggests weaknesses in the Mr. Simiyu of the KARI is highly systems of information flow, both from appreciated. The whitefly and research through extension to farmers parasitoid specimens were mounted and from farmers back to research. and identified by Dr. Mohammed Ali Bob of the International Center of During the survey conducted in Insect Physiology and Ecology. this study, the impact of the pandemic of severe CMD has been captured both through the description of the References unusually high incidence and severity of the disease in Western Province and Bock, K. R. 1975. Maize streak, cassava through the estimates of major yield latent and similar viruses. In: Abstracts, Third International losses provided by farmers interviewed. Congress for Virology, Madrid, ES. Considerable experience in tackling p. 93. this problem has been developed

58 Kenya

Bock, K. R. 1983. Epidemiology of Harrison, B. D. 1987. Properties and cassava mosaic disease in Kenya. geographical variation of In: Plumb, R. T.; Thresh, J. M. geminivirus isolates from mosaic- (eds.). Plant virus epidemiology. affected cassava. In: African Blackwell Scientific Publications, cassava mosaic disease and its Oxford, GB. p. 337-347. control. Procs. International Seminar, 4-8 May 1987, Bock, K. R. 1994a. Control of African Yamoussoukro, CI. Centre cassava mosaic geminivirus by Technique de Coopération Agricole using virus-free planting material. et Rurale, Wageningen, NL, and Trop. Sci. 34:102-109. Institut Français de Recherche Scientifique pour le Dévéloppement Bock, K. R. 1994b. The spread of African en Coopération, Paris, FR. p. 270. cassava mosaic geminivirus in coastal and western Kenya. Trop. Harrison, B. D.; Zhou, X.; Otim-Nape, G. Sci. 34:92-101. W.; Liu, Y.; Robinson, D. J. 1997. Role of a novel type of double Bock, K. R.; Woods, R. D. 1983. Etiology infection in the geminivirus- of African cassava mosaic disease. induced epidemic of severe cassava Plant Dis. 67:944-955. mosaic in Uganda. Ann. Appl. Biol. 131:437-448. Carey, E. E.; Mwanga, R. O. M.; Fuentes, S.; Kasule, S.; Macharia, C.; Legg, J. P. 1999. Emergence, spread and Gichuki, S. T.; Gibson, R. W. 1998. strategies for controlling the Sweetpotato viruses in Uganda and pandemic of cassava mosaic virus Kenya: Results of a survey. In: disease in east and central Africa. Procs. Sixth Triennial Symposium of Crop Prot. 18:627-637. the International Society of Tropical Root Crops-Africa Branch (ISTRC- Legg, J. P.; Kapinga, R.; Teri, J.; Whyte, J. AB), 22-28 October 1995, Lilongwe, B. A. 1999. The pandemic of MW. p. 457-461. cassava mosaic virus disease in East Africa: Control strategies and Doughty, L. R. 1958. Cassava breeding for regional partnerships. Roots resistance to mosaic and brown 6(1):10-19. streak viruses. In: Annual report. East Afr. Agric. For. Res. p. 48-51. Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; Baguma, Y.; Ogwal, S.; Semakula, Gibson, R. W.; Mpembe, I.; Alicai, T.; G. N.; Acola, G.; Byabakama, B.; Carey, E. E.; Mwanga, R. O. M.; Martin, A. 1997. Cassava mosaic Seal, S. E.; Vetten, H. J. 1998. virus disease in Uganda: The Symptoms, aetiology and serological current pandemic and approaches analysis of sweetpotato virus to control. Natural Resources disease in Uganda. Plant Pathol. Institute (NRI), Chatham, GB. 65 p. 47:95-102. Robertson, I. A. D. 1987. The role of Gregory, P. H. 1948. The multiple Bemisia tabaci (Gennadius) in the infection transformation. Ann. Appl. epidemiology of ACMV in East Biol. 35:412-417. Africa: Biology, population dynamics and interaction with cassava varieties. In: African cassava mosaic disease and its control. Procs. International Seminar, 4-8 May 1987, Yamoussoukro, CI. Centre Technique de Coopération Agricole et Rurale, Wageningen, NL, and Institut Français de Recherche Scientifique pour le Dévéloppement en Coopération, Paris, FR. p. 57-63.

59 Whiteflies and Whitefly-borne Viruses in the Tropics

Seif, A .A. 1981. Seasonal fluctuation of Sheffield, F. M. L. 1957. Virus diseases of adult population of the whitefly, sweetpotato in East Africa. 1. Bemisia tabaci, on cassava. Insect Identification of viruses and their Sci. Appl. 1:363-364. insect vectors. Phytopathology 47:582-587. Seif, A. A. 1982. Effect of cassava mosaic virus on yield of cassava. Plant Dis. 66:661-662.

60 Tanzania

CHAPTER 1.8 Tanzania

Joseph Ndunguru*, Simon Jeremiah*, Regina Kapinga** and Peter Sseruwagi***

Introduction recorded as causing serious losses until the 1920s. Between 1920 and Cassava (Manihot esculenta Crantz) and 1960, comprehensive studies were sweetpotato (Ipomoea batatas [L] Lam.) conducted in the country, emphasizing are important staple food crops, the development of CMD-resistant especially in the rural communities of varieties through a breeding program Tanzania. The two crops have a long conducted at Amani in the Usambara history of providing food security in the Mountains (Jennings, 1994). Resistant country, particularly during famine. varieties developed by the programme However, their production is currently were effective in controlling CMD and threatened by cassava mosaic disease restored the crop’s productivity. Data (CMD) and sweetpotato virus disease obtained in 1989 and 1990 during the (SPVD). CMD is caused by cassava first phase of the Collaborative Study of mosaic begomoviruses (CMBs) Cassava in Africa (COSCA) indicated transmitted by the whitefly Bemisia that Tanzania had the lowest CMD- tabaci (Gennadius) and through virus- incidence (37%) of the six countries infected planting material (Harrison, surveyed (Thresh et al., 1994). Another 1987). On the other hand, SPVD extensive survey, conducted between results from co-infection by two distinct 1993 and 1994 in mainland Tanzania viruses, Sweetpotato chlorotic stunt virus and the islands of Zanzibar and transmitted by B. tabaci and Pemba, rated the incidence of CMD at Sweetpotato feathery mottle virus 28%, with infected cuttings (24%) transmitted by the aphid Myzus providing the major source of infection persicae (Sulzer) (Gibson et al., 1998). (Legg and Raya, 1998). During the latter part of the 1990s, CMD research CMD was first reported in Tanzania has been reinvigorated and has under the name “Krauselkrankheit” included work to assess the (Warburg, 1894), although it was not development of the disease in the country.

* Root and Tubers Programme of the Lake Recent epidemiological studies on Zone Agricultural Research and CMD elsewhere in East Africa have Development Institute (LZARDI), Ukiriguru, provided evidence of the north-to-south Mwanza, Tanzania. ** Centro Internacional de la Papa (CIP), spread of an epidemic of severe CMD, Kampala, Uganda. firstly within Uganda (Otim-Nape et al., *** International Institute of Tropical 1997; Legg and Ogwal, 1998; Chapter Agriculture-Eastern and Southern Africa Regional Center (IITA-ESARC), Kampala, 1.6, this volume) and subsequently into Uganda. parts of western Kenya and Tanzania

61 Whiteflies and Whitefly-borne Viruses in the Tropics bordering Uganda (Legg, 1999). This expansion has been associated with Uganda Lake Victoria the spread of a novel CMB variant Kenya Kagera (Harrison et al., 1997). The pandemic Rwanda Mara poses an immediate threat to cassava Mwanza Burundi production in the areas of Tanzania Shinyanga bordering Uganda and eventually to the country as a whole. Research towards Tanga controlling the disease is the focus of Tanzania continuing work by the Root and Dar-es-Salaam Tubers Programme of the Lake Zone Pwani Agricultural Research and Development Institute (LZARDI) based at Ukiriguru, Lindi Mwanza. The institute is supported in this effort by other international D.R. Congo Mtwara research institutions and networks, including the International Institute of Mozambique Tropical Agriculture-Eastern and Southern Africa Regional Center Figure 1. Cassava- and sweetpotato-growing (IITA-ESARC), the East African Root areas surveyed for whitefly incidence Crops Research Network (EARRNET) in Tanzania. and the Southern African Root Crops Research Network (SARRNET). SPVD, Increased Biological and its effects on the production of Understanding sweetpotato, has received less attention in Tanzania. However, the disease poses a threat to the production of the Whitefly species and crop if left unchecked. abundance

Diagnostic surveys were conducted Three whitefly species were identified in mid-1998. Three target areas were on cassava: B. tabaci, B. afer (Priesner chosen to represent the major cassava and Hosny) and Trialeurodes ricini and sweetpotato growing areas in the (Misra). Only B. tabaci and B. afer were country: the “lake zone” (i.e., bordering identified on sweetpotato. Of Lake Victoria), which includes Mwanza, 303 whitefly samples collected on Shinyanga, Mara and Kagera regions; cassava, 175 were B. tabaci, the “southern coast”, which includes 127 B. afer and one T. ricini. The Lindi and Mtwara regions; and the 105 whitefly samples from sweetpotato “northern coast”, which includes comprised 89 B. tabaci and 16 B. afer. Tanga, Dar-es-Salaam and Pwani The species B. tabaci and B. afer were regions (Figure 1). The study aimed at found in all three survey areas. B. afer identifying whiteflies and whitefly- was more common on cassava than on transmitted viruses and characterizing sweetpotato. It comprised 26% of the producer knowledge and responses to whiteflies in samples collected on these problems on cassava and cassava and 9% of those on sweetpotato in Tanzania. sweetpotato from the lake zone, 53% on cassava and 11% on sweetpotato from the southern coast, and 45% on cassava and 27% on sweetpotato from the northern coast (see also Figure 2, Chapter 1.14, this volume).

62 Tanzania

Adult whiteflies were more 19 in the northern coast samples, abundant on cassava in the northern while only one sample of Encarsia sp. coast (5.6 per top five leaves) than in (luteola group) was recorded, in the either the southern coast (1.1) or the northern coast. Little is known about lake zone (0.7). On sweetpotato, the role of natural enemies in the estimates of whitefly abundance were population dynamics of whiteflies in highest in the lake zone (40.1 per Tanzania. 1-min count) and lowest in the southern coast (2.8) Increased Socio-Economic Disease incidence and Understanding symptom severity CMD incidence was generally low in the Farmers’ assessment of three target areas. The highest whitefly-related problems incidence was recorded in the northern coast (33.6%) and the lowest (9.4%) in A small proportion of both cassava the southern coast (Table 1). The main (30%) and sweetpotato (11.7%) farmers source of infection in the lake zone were able to recognize whiteflies on the (69.3% of the total infection in area) two crops, and even the names given to and northern coast (56.8% of the total the insect were non-specific. Moreover, infection in area) was cutting infection. of the farmers who recognized the However, whitefly-borne infection was whiteflies only 16.7% (on cassava) and the predominant source of infection in 3.3% (on sweetpotato) considered them the southern coast. CMD symptoms a problem in the production of their were mild in the lake zone and respective crops. southern coast and slightly more severe in the northern coast. Most cassava farmers (70%) could recognize CMD as a disease of cassava, The incidence of SPVD was highest 58% recognized it as a constraint to in the southern coast (12.4%), followed crop production and 68% as a problem by the lake zone (11.3%), and lowest in occurring every year in their crops. In the northern coast (4.2%) survey area contrast, only 38% of sweetpotato (Table 1). SPVD symptoms were farmers recognized SPVD as a disease relatively mild in most of the surveyed of sweetpotato, 32% recognized it as a regions, with the exception of constraint to production and 41% as a Shinyanga region in the lake zone, yearly recurring problem. One third of where the disease was very severe (4.5). cassava farmers considered that CMD The low incidence of SPVD should severity was increasing, whilst a enable the use of phytosanitary slightly smaller proportion of measures, including selection of sweetpotato farmers considered that planting material and roguing in the SPVD was becoming more important. management of the disease. Forty percent of cassava farmers and 32% of sweetpotato farmers believed Whitefly parasitoids that climate influenced the diseases; two-thirds believed that they were more The survey identified two whitefly prevalent during periods of low rainfall parasitoids, Encarsia sophia (Girault and high temperatures. Cassava and Dodd) and Encarsia sp. (luteola farmers attributed higher losses to group). E. sophia was by far the more CMD than did sweetpotato farmers to widely recorded, with 43 samples SPVD. Using five categories (zero, identified in the southern coast and quarter, half, three quarters and total

63 Whiteflies and Whitefly-borne Viruses in the Tropics SPVD a dance on cassava and sweet Sweet potato and on sweet potato (per minute gory, P. H. 1948. The multiple No. Whitefly 0 2.8 2.8 9 2.9 8.3 2.1 a Whitefly Cutting Total Severity Incidence Severity infection infection incidence 15.4 (18.0)15.4 6.7 22.1 3.0 13 1.8 9.2 2.7 3 4.3(16.3) 14.4 4.4 18.8 3.0 5 5.0 3.3 2.0 No. Whitefly CMD fields counts fields counts b Ann. Appl. Biol. 35:412-417); severity of disease measured on an ascending 1-5 scale, from low to severe. . Region Cassava MaraShinyangaMwanza 5 Mean 5Lindi 4Mtwara 0.4 Mean 0.1 14Pwani 6 0.2(0.9) 0.7 DSM (1.4) 1.3 0.8 3.3 (4.1) 7 20.7 1.2 1.0 7.3(1.7) 1.2 1.2 18.0(9.2) 8.1 (2.8) 2.8 21.4 8.9 8.6 19.8(7.1) 6.5 21.3 7.9 2.6 21.0 2.6 5.5 2.4 16.0 2.6 4 12.0 6 2.5 5 2.7 25.3 55.9 11 61.1 2.0 5.3 40.1 2.7 18.0 4.5 2.8 11.5 2.8 16.4 3.1 3.3 12.8 2.4 2.3 Tanga Mean 10 3.6 5.8(29.5) 13.7 (21.0) 14.4 46.3 26.2 60.0 40.6 3.1 3.0 2 5.7 3.1 0 6.8 2.0 2.5 potato in selected regions of Tanzania, 1998. count); whitefly infection, figures in parentheses transformed to multiple infection units allow for (Gre infection transformation Survey area Lake zone KageraSouthern coast 6Northern coast 2.2 0 (0) 30.7 30.7 2.6 5 18.0 20.0 3.1 b. DSM, Dar es Salaam. Table 1. Incidence (%) of cassava mosaic disease (CMD) and sweet potato virus (SPVD), severity whitefly abun a. Figures are means for each region. Whitefly counts, whitefly abundance on cassava (number of whiteflies per top five leaves)

64 Tanzania loss) to assess yield loss attributable to material. The use of host plant either CMD or SPVD, 20% of the resistance as a management option was cassava farmers believed that CMD mentioned by only 2% of cassava causes at least 50% yield loss, while farmers but not mentioned at all by 23% considered the disease to cause sweetpotato farmers. Only 30% of only 25% yield loss. Among the cassava farmers and 22% of sweetpotato farmers, over 50% sweetpotato farmers noted differences considered that SPVD caused only in response to CMD and SPVD between 25% yield loss. varieties. Those that did so mainly attributed it to differences in levels of Managing whiteflies and resistance to disease. In general, very whitefly-transmitted viruses few cassava (8%) and sweetpotato (2%) farmers had received any technical Both cassava and sweetpotato farmers assistance in the management of attempted to control CMD and SPVD whiteflies and associated virus diseases by roguing and selecting disease-free on either crop. However, most farmers planting material. Roguing was used were willing to change the planting for disease management by 23% of dates of their crop and monitor whitefly cassava farmers and 5% of sweetpotato and disease problems if it would help in farmers. Farmers growing cassava gave the management of either disease. three reasons for roguing diseased plants: reducing the spread of the disease (18%), poor growth (3%) and preventing disease (2%). However, only Conclusions 5% of sweetpotato farmers considered Both diseases covered in this study are that roguing would reduce SPVD. Most becoming more prevalent in Tanzania cassava farmers rogued infected plants (Chapters 1.13 and 1.14, this volume) before the crop was 4 months old. For and therefore measures to reduce their sweetpotato, there was no specific impact are needed urgently. period during which roguing was done. Seventy-three percent of cassava farmers compared with 47% of Farmers are more aware of CMD sweetpotato farmers selected clean than of SPVD. However, their planting material, with absence of knowledge of management options for disease symptoms as the selection the two diseases is weak, implying poor criterion for the two diseases. In both communication among researchers, cases, however, farmers considered the extension workers and farmers. two principal methods to be only Farmers’ knowledge of these diseases partially effective in managing the should be enhanced through diseases. participatory training, while researchers, farmers and other Only one sweetpotato farmer and stakeholders should work together to none of the cassava farmers mentioned develop and validate cost-effective chemical control as a management disease and vector management tactic. The sweetpotato farmer used the strategies and encourage their wider organophosphate pesticide dimethoate, adoption by producers. applied 3 times, when whiteflies and damage were observed in the field. Few Varieties that are resistant to CMD farmers reported abandoning the and SPVD, acceptable to farmers and production of cassava (10%) and well adapted to local conditions are sweetpotato (3%) but many reported urgently needed. The development, occasional shortage of clean planting farm-level evaluation and wider

65 Whiteflies and Whitefly-borne Viruses in the Tropics distribution of improved varieties of References cassava and sweetpotato should be strengthened and expanded. Gibson, R. W., Mpembe, I.; Alicai, T.; Carey, E. E.; Mwanga, R. O. M.; Seal, S. E.; Vetten, H. J. 1998. The whitefly parasitoid E. sophia Symptoms, aetiology and occurred frequently in the areas serological analysis of sweetpotato surveyed, and studies elsewhere in virus disease in Uganda. Plant East Africa have shown that parasitoids Pathol. 47:95-102. can cause B. tabaci nymph mortality rates of up to 50%. The deployment of Gregory, P. H. 1948. The multiple resistant varieties and phytosanitation infection transformation. Ann. Appl. Biol. 35:412-417. measures probably will be the most readily applicable CMD and SPVD Harrison, B. D. 1987. Properties and management measures in Tanzania. geographical variation of However, there may be scope in the geminivirus isolates from mosaic- future for improved management of the affected cassava. In: African whitefly vector through the cassava mosaic disease and its control, Procs. International development and application of Seminar, 4-8 May 1987, measures designed to conserve and Yamoussoukro, CI. Centre augment the activity of whitefly natural Technique de Coopération Agricole enemies. The farmer field school (FFS) et Rurale, Wageningen, NL, and approach is being widely used in Centre Technique de Coopération Tanzania, and FFS in root crop growing Agricole et Rurale, Wageningen, NL areas may provide an ideal opportunity and Institut Français de Recherche Scientifique pour le Dévéloppement to strengthen farmers’ knowledge of en Coopération, Paris, FR. p. 270. CMD and SPVD and develop a holistic, integrated approach to their Harrison, B. D.; Zhou, X.; Otim-Nape, G. management, incorporating all possible W.; Liu, Y.; Robinson, D. J. 1997. components including host plant Role of a novel type of double resistance, cultural methods and infection in the geminivirus- biological control. induced epidemic of severe cassava mosaic in Uganda. Ann. Appl. Biol. 131:437-448.

Acknowledgements Jennings, D. L. 1994. Breeding for resistance to African cassava The technical support of Dr. Richard mosaic geminivirus in East Africa. Gibson of the Natural Resources Trop. Sci. 34:110-122. Institute, UK, Mr. Mathew Raya, Mr. Geoffrey Mkamiro, Mr. Raphael Legg, J. P. 1999. Emergence, spread and strategies for controlling the Msabaha of Naliendele Agricultural pandemic of cassava mosaic virus Research Institute and Ms. Hellen disease in east and central Africa. Kiozya and Ms. Esther Masumba of Crop Prot. 18:627-637. Kibaha Research Institute is highly appreciated. The authors also Legg, J. P.; Ogwal, S. 1998. Changes in acknowledge Dr. Mohammed Ali Bob of the incidence of African cassava the International Center of Insect mosaic geminivirus and the abundance of its whitefly vector Physiology and Ecology for his along south-north transects in identification of whitefly species and Uganda. J. Appl. Entomol. natural enemies. 122:169-178.

66 Tanzania

Legg, J. P.; Raya, M. 1998. Survey of Thresh, J. M.; Fishpool, L. D. C.; Otim- cassava virus diseases in Nape, W.; Fargette, D. 1994. Tanzania. Int. J. Pest Manage. African cassava mosaic virus 44:17-23. disease: An under-estimated and unsolved problem. Trop. Sci. Otim-Nape, G. W.; Bua, A.; Thresh, J. 34:3-14. M.; Baguma, Y.; Ogwal, S.; Semakula, G. N.; Acola, G.; Warburg, O. 1894. Die Kulturpflanzen Byabakama, B.; Martin, A. 1997. Usambaras. Mitt. Dtsch. Schutz. Cassava mosaic virus disease in 7:131. Uganda: The current pandemic and approaches to control. Natural Resources Institute (NRI), Chatham, GB. 65 p.

67 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.9 Malawi

Mathews P. K. J. Theu* and Peter Sseruwagi**

Introduction

Cassava mosaic disease (CMD) is caused by cassava mosaic Tanzania begomoviruses, transmitted by the whitefly Bemisia tabaci (Gennadius) Rumphi and through virus-infected planting Zambia material (Harrison, 1987). The Nkhata disease has gained significance as a Bay major constraint to production in many areas of Malawi where cassava Nkhotakota Mozambique (Manihot esculenta Crantz) is grown, Dowa especially in the lowlands, where Salima conditions are warmer; infection Lilongwe reportedly exceeds 90% in some Dedza plantings (Nyirenda et al., 1993).

Malawi A survey was conducted in three target areas selected to represent the major cassava growing areas in Malawi: the central lakeshore Zimbabwe (Salima and Nkhotakhota, at 500-800 m altitude); the northern Figure 1. Areas surveyed for whitefly incidence lakeshore (Rumphi and Nkhata Bay, and cassava mosaic disease in 500-800 m); and the central plateau Malawi. (Dedza, Lilongwe and Dowa, 1000-1500 m) (Figure 1). The key characterize producer knowledge of objectives of the study were to these problems on cassava in Malawi. identify whiteflies and whitefly- Although assessments of sweetpotato transmitted viruses and to virus disease (SPVD) also were planned initially, surveys were done at a time (July-September 1998) when crops of * Chitedze Research Station, Lilongwe, Malawi. sweetpotato (Ipomoea batatas [L.] Lam.) ** International Institute of Tropical mostly had been harvested, so disease Agriculture-Eastern and Southern Africa Regional Center (IITA-ESARC), Kampala, and vector incidence could not be Uganda. assessed adequately.

68 Malawi

Increased Biological Whitefly parasitoids Understanding Only one species of whitefly parasitoid, Encarsia sophia (Girault and Dodd) was identified and this was from the Whitefly species and northern lakeshore. abundance Two whitefly species, B. tabaci and B. afer (Priesner and Hosny), were Increased Socio-Economic identified on cassava. Altogether, Understanding 212 whitefly samples were collected on the crop. Of these, 65 were B. tabaci and 147 were B. afer. The relative Farmers’ assessment of abundance of the two species varied whitefly-related problems among survey areas. B. afer was Forty-four percent of the farmers especially abundant in the central mid- interviewed were able to recognize altitude plateau, where it comprised whiteflies on the crop but only 32% 97% of the whitefly samples collected. considered them a problem. Most The adult whitefly population did not farmers (61%) were able to recognize vary significantly among the regions CMD as a disease of cassava and gave surveyed (Table 1). it a range of local names. Less than half the farmers (42%) recognized the Disease incidence and disease as a problem to cassava symptom severity production and most (64%) noted that The incidence of CMD ranged from it occurs yearly. Only 22% believed 11.4% in the central plateau survey that CMD was becoming more severe, area to 62.2% for the central lakeshore with 1997 noted as the year of most (Table 1). The use of diseased cuttings severe disease symptoms. About one was the main source of infection in all quarter (27%) of the farmers believed three survey areas. Symptoms of CMD the disease to be affected by climate. were moderately severe in each of the There was no consensus on the effect survey areas and averaged 2.8 overall. of rainfall on CMD but 39% of farmers

Table 1. Cassava mosaic disease (CMD) incidence (%), disease symptoms and whitefly counts in selected areas of Malawi, 1998. Target area Provinces/ No. Whitefly CMDa districts fields counts Whitefly Cutting Total Severity infection infection incidence Central Salima and lakeshore Nkhotakota 20 1.5 27.5 (54.7) 34.7 62.2 2.8

Northern Rumphi and lakeshore Nkhata Bay 9 1.0 19.6 (34.4) 32.6 52.2 2.7

Central Dedza, Lilongwe plateau and Dowa 12 1.3 3.3 (3.7) 8.1 11.4 3.0

Mean 1.3 16.8 (25.4) 25.1 41.9 2.8 a. Figures are means for each region. Whitefly counts, whitefly abundance on cassava (number of whiteflies per top five leaves); whitefly infection, figures in parentheses transformed to multiple infection units to allow for multiple infection (Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. Biol. 35:412-417); severity of disease measured on an ascending 1-5 scale, from low to severe.

69 Whiteflies and Whitefly-borne Viruses in the Tropics considered that the disease was most Conclusions severe during periods of high temperature. Farmers’ estimates of the Prospects for managing CMD losses attributable to CMD varied successfully are good in most locations considerably, ranging from “none” to sampled, since the use of diseased “total crop loss”. cuttings was the main source of infection. However, despite the majority Managing whiteflies and of farmers recognizing the disease as a whitefly-transmitted viruses production constraint to cassava, their knowledge of the disease was weak. Only 7.3% of farmers attempted to The incidence of CMD was relatively control CMD; 2.4% rogued, 3.7% high in the lakeshore survey areas and, selected for healthy planting material given that moderate symptoms were at planting, while some 1.2% observed, it is likely that significant de-topped CMD-infected plants to yield losses are being experienced. reduce plant shoots with disease Given that there appears to be very symptoms in the field. Farmers who limited CMD spread due to the activity selected their planting material mainly of whiteflies, and that B. afer, a non- used the absence of disease symptoms vector, is the predominant cassava as the selection criterion. Farmers who whitefly species in Malawi, rogued gave three reasons for doing phytosanitation may provide an so—prevention of disease spread (7%), effective approach to CMD reduction of disease (7%) and removal management. However, before an of poorly growing plants (5%). Most of appropriate regime for the use of the roguing was done before the crop roguing and the selection of CMD-free was 4 months old. Only 5% of farmers planting material can be proposed, considered selection and 7% basic information will be required on considered roguing as highly effective yield losses and rates of infection in the in controlling the disease. Some major cassava-growing areas for the farmers (10%) reported the occasional most commonly grown varieties. These shortage of clean planting material, results should be used to develop a with 5% reporting abandoning programme of participatory evaluation production of the crop in 1990 and with farmers of both phytosanitation 1992. This may have been associated and new varieties. This should be with the severe damage caused by the closely linked with an educational cassava mealybug, Phenacoccus programme targeted at increasing manihoti Matile-Ferrero, at this time. farmers’ knowledge of CMD and the Improved varieties, primarily developed whitefly vector and of the range of for their superior yield and agronomic possible management approaches. characteristics, were grown by only 5% of the farmers, who considered their performance to be better than that of Acknowledgements local varieties. Very few farmers (5%) had received any technical assistance The authors gratefully recognize the in the management of whiteflies. technical support of Mr. McDonald However, a large proportion (40%) of Kavala, Mr. Elias Kamsikili and farmers were willing to monitor whitefly Mr. Eric Mazuma. Assistance from and disease problems if it would help Dr. Mohammed Ali Bob of the in the management of the disease. International Center of Insect

70 Malawi

Physiology and Ecology in the Nyirenda, G. K. C.; Munthali, D. C; Phiri, identification of whitefly species and G. S. N.; Sauti, R. F. N.; Gerling, D. natural enemies was highly 1993. Integrated pest management of Bemisia spp. whiteflies in Malawi. appreciated. Makoka Research Station, Thondwe, MW. References

Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. Biol. 35:412-417.

Harrison, B. D. 1987. Properties and geographical variation of geminivirus isolates from mosaic- affected cassava. In: African cassava mosaic disease and its control. Procs. International Seminar, 4-8 May 1987, Yamoussoukro, CI. Centre Technique de Coopération Agricole et Rurale, Wageningen, NL, and Institut Français de Recherche Scientifique pour le Dévéloppement en Coopération, Paris, FR. p. 270.

71 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.10 Madagascar

Sahondramalala Ranomenjanahary*, Jocelyn Ramelison* and Peter Sseruwagi**

Introduction become a problem of relatively minor significance. Cassava mosaic disease (CMD), caused by cassava mosaic begomoviruses Sweetpotato virus disease (SPVD) transmitted by the whitefly Bemisia results from co-infection of sweetpotato tabaci (Gennadius) and through virus- (Ipomoea batatas [L.] Lam.) by whitefly- infected planting material (Harrison, transmitted Sweetpotato chlorotic stunt 1987), is the most important viral virus and aphid-transmitted disease of cassava (Manihot esculenta Sweetpotato feathery mottle virus Crantz) in Madagascar. The disease (Gibson et al., 1998). SPVD has not was first reported in Madagascar in been reported as a major problem in 1932, although it was of only minor Madagascar, although the country is importance at this time (François, one of the largest producers of 1937). Epidemics were first reported in sweetpotato in Africa. 1934, from an area to the west of Lake Alatroa on the central plateau and, In order to provide a basic between this time and the end of the understanding of the incidence of 1930s, the epidemic spread to cover all whiteflies, CMD and SPVD in cassava-growing areas of the island Madagascar, a study was conducted in (Cours, 1951). Symptoms in local four survey areas that represent the varieties were very severe and cassava major cassava and sweetpotato growing cultivation was abandoned on a wide areas: Antananarivo (central mid- scale. The impact was so great that the altitude plateau), Fianarantsoa (eastern government ordered urgent attention to humid coast), Toliara (south-western be given to the problem (Frappa, 1938) semi-arid coast) and Mahajanga and a major programme for the (northern sub-humid coast) (Figure 1). development of resistant varieties was This initial phase of the study aimed at initiated (Cours, 1951; Cours-Darne, documenting the status of whiteflies 1968). The first of these varieties was and whitefly-transmitted viruses and at widely disseminated in the mid-1940s. describing producer knowledge of these Since this time, CMD appears to have problems on cassava and sweetpotato, with a view to prioritizing further actions and eventually developing * Laboratoire de Pathologie Végétale, integrated management strategies to Ambatobe, Madagascar. tackle the most serious problems ** International Institute of Tropical identified. The survey was carried out Agriculture-Eastern and Southern Africa Regional Center (IITA-ESARC), Kampala, during the months of June and August Uganda. 1998.

72 Madagascar

the lowest (31%) was recorded in Antananarivo (Table 1). The planting of infected cuttings was much more important than whitefly transmission Tanzania Madagascar as a source of infection and comprised 86% of total infection. CMD symptom severity was relatively mild in Antananarivo but was severe Mahajanga elsewhere.

SPVD was of very low incidence or Antananarivo absent, except in the Toliara survey Mozambique area, where an incidence of 18% was recorded. SPVD symptoms were severe in Antananarivo but very mild where Fianarantsoa the disease occurred elsewhere.

Toliara Increased Socio-Economic Understanding

Farmers’ assessment of Figure 1. Cassava- and sweetpotato-growing whitefly-related problems areas surveyed for whitefly incidence in Madagascar. Thirteen percent of cassava farmers but only 3% of sweetpotato farmers recognized whiteflies. Scarcely any of Increased Biological these farmers considered whiteflies to Understanding be a problem on either crop. Almost one fifth (19%) of cassava farmers were able to recognize CMD as a disease of Whitefly species and cassava and gave the disease a range of abundance local names, whereas no sweetpotato farmers recognized SPVD. Only 6% of Two whitefly species, B. tabaci and cassava farmers considered CMD as a B. afer (Priesner and Hosny) were production constraint to cassava. They identified on both cassava and noted that the disease occurs yearly sweetpotato. Of 415 whitefly samples and that it was becoming more severe. collected on the two crops, 118 were Climate affects CMD and SPVD B. tabaci and 297 were B. afer. Adult incidence according to 30% of cassava whitefly populations on cassava were farmers and 5% of sweetpotato greatest in the Antananarivo survey farmers. In particular, 19% of cassava area and lowest in Mahajanga (Table 1, farmers noted that disease is more see also Figure 3 in Chapter 1.14, this severe during periods of little rain and volume). Whiteflies were very scarce on high temperature but sweetpotato sweetpotato. farmers did not report this effect. None of the farmers could estimate the level Disease incidence and of yield loss attributable to either symptom severity disease. The lack of recognition of The highest disease incidence (71%) CMD, first as a disease and second as was recorded in Fianarantsoa, while a constraint to production, is

73 Whiteflies and Whitefly-borne Viruses in the Tropics SPVD a on cassava and Sweetpotato ) and on sweetpotato (per minute gory, P. H. 1948. The multiple No. Whitefly fields counts a (7.8) 25.5 31.0 2.6 11 0.01 2.6 3.3 (11.0) 34.0 41.0 3.3 6 0 0 - Whitefly Cutting Total Severity Incidence Severity infection infection incidence Ann. Appl. Biol. 35:412-417); severity of disease measured on an ascending 1-5 scale, from low to severe. . No. Whitefly CMD fields counts sweetpotato in selected regions of Madagascar, 1998. count); whitefly infection, figures in parentheses transformed to multiple infection units allow for (Gre infection transformation Province Cassava AntananarivoFianarantsoa 35Toliara 22Mahajanga Mean 7.1 42 12 4.3 2.5 5.6 5.9 5.0(21.9) 7.1 6.9 6.4 (11.2) 63.9(11.6) 6.5 71.0 39.7 40.8 46.0 3.5 47.3 15 3.2 3.1 7 - 0 0.10 2.4 0.03 17.6 5.7 2.0 2.3 2.5 Table 1. Incidence of cassava mosaic disease (CMD) and sweetpotato virus (SPVD), severity whitefly abundance a. Figures are means for each province. Whitefly counts, whitefly abundance on cassava (number of whiteflies per top five leaves

74 Madagascar surprising, particularly given the mainland, at least in three of the four history of previous CMD epidemics in survey areas. Infection was mainly Madagascar and the severity of the attributable to the use of diseased disease symptoms observed. planting material. However, the high severity of the symptoms in plantings Managing whiteflies and where the disease does occur suggests whitefly-transmitted viruses that farmers are probably sustaining significant yield losses. It would be Selection of clean planting material was valuable to quantify yield losses in local the only disease management approach cultivars being grown in each of the used—as reported by 40% of cassava target areas. Such an exercise should farmers and 13% of sweetpotato be carried out with the full farmers. Sixteen percent of cassava participation of farmers and could be farmers and 3% of sweetpotato farmers used simultaneously to introduce new used the absence of disease symptoms germplasm and evaluate its as selection criterion but both performance and acceptability. considered the method to be only partially effective. Only 27% of cassava farmers and 3% of sweetpotato farmers The sharp disparity in SPVD were able to recognize differences incidence between Toliara and the between varieties of their respective other three survey areas is difficult to crops and even fewer (only 2% of explain. Although the abundance of cassava farmers and no sweetpotato whitefly adults was greatest in this farmers) could describe or characterize target area, whitefly populations were these differences. Cassava farmers generally very low compared to those growing improved varieties noted that recorded in other countries. In their performance was only slightly addition, at the time the survey was better than the local varieties. Only 4% conducted, B. afer, a non-vector, was of cassava farmers and no sweetpotato the predominant species on both farmers received technical assistance cassava and sweetpotato. More detailed in whitefly or disease management. epidemiology and population dynamics None of the farmers considered that studies investigating changes in environmental factors such as certain abundance of the two whitefly species patterns of weather gave them any idea and disease spread over time would be about what whitefly populations or required in order to identify the disease levels would result and most reason(s) behind this difference. were unwilling to change the planting dates of their crops even if it might Both cassava and sweetpotato help reduce the effects of whiteflies farmers had inadequate knowledge of and/or disease. The farmers’ the principal diseases and the options reluctance is understandable, however, available for managing them. A given the relatively short rainy season participatory research and training in much of Madagascar that leaves effort would enhance farmers’ them little scope to adjust planting knowledge and would help researchers dates. understand the constraints to farmers’ adoption of resistant varieties and other potential management tactics. It Conclusions would be valuable also to continue to monitor the incidence of the two The incidence of CMD was low in many diseases in the country and strengthen farmers’ fields, compared with figures the national breeding program, to reported from countries on the African compare local sources of resistance

75 Whiteflies and Whitefly-borne Viruses in the Tropics with those available elsewhere and so François, E. 1937. Un grave peril: La develop resistant varieties acceptable to mosaïque du manioc. Agron. Colon. farmers. 26:33-38. Frappa, C. 1938. Les insectes nuisibles au manioc sur pied et aux Acknowledgements tubercules de manioc en magasin à Madagascar. Rev. Bot. Appl. 28: Appreciation is extended to Ms. Bakoly 17-29. Ravoniarimelina, Centre National de Recherche sur l’Environnement (CNRE) Gibson, R. W.; Mpembe, I.; Alicai, T.; and Mr. Ernest Randrianarisoa, Carey, E. E.; Mwanga, R. O. M.; Seal, S. E.; Vetten, H. J. 1998. Departement d’Entomologie, Faculté Symptoms, aetiology and serological des Sciences, Université analysis of sweetpotato virus d’Antananarivo, who participated in disease in Uganda. Plant Pathol. data collection, and Dr. Mohammed Ali 47:95-102. Bob, of the International Center of Insect Physiology and Ecology, for his Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. identification of whitefly species. Biol. 35:412-417.

Harrison, B. D. 1987. Properties and References geographical variation of geminivirus isolates from mosaic- Cours, G. 1951. Le manioc à Madagascar. affected cassava. In: African cassava Mémoires de l’Institut Scientifique mosaic disease and its control. de Madagascar Série B Biol. Vég. Procs. International Seminar, 3:203-400. 4-8 May 1987, Yamoussoukro, CI. Centre Technique de Coopération Cours-Darne, G. 1968. Improving cassava Agricole et Rurale, Wageningen, NL, in Africa. The Abidjan Conference. and Institut Français de Recherche Agr. Res. Prior. Econ. Dev. Afr. Scientifique pour le Développement 2:330-339. en Coopération, Paris, FR. p. 270.

76 Diversity of African Cassava Mosaic Disease

CHAPTER 1.11 The Diversity of Cassava Mosaic Begomoviruses in Africa

Peter Markham*, Robert Briddon*, Clotilde Roussot*, John Farquhar*, Geoffrey Okao-Okuja** and James Legg***

Introduction whitefly Bemisia tabaci (Gennadius). For some time, ACMV was thought to Cassava mosaic disease (CMD) is the be the only geminivirus species most widespread and economically associated with CMD, despite the important disease of cassava (Manihot suggestion by Harrison and Robinson esculenta Crantz) in tropical Africa. A (1988), based upon serological disease of cassava was first described analysis, that CMD was regionally in East Africa in the nineteenth century distinct. Some 10 years after the (Warburg, 1894) and cassava mosaic, publication of the sequence of ACMV, as it came to be called, was Hong et al. (1993) characterized a subsequently reported as spreading further CMD-associated geminivirus. throughout the cassava-growing areas Although based upon only the of Central and West Africa (Calvert and sequence of the DNA A genomic Thresh, 2002). In more recent times, component (and therefore lacking any CMD has been reported as occurring at corroborative information on infectivity, varying levels of incidence throughout since this is determined by the the cassava belt of Africa and losses B genome component) this virus was have been estimated at between 12% given the name East African cassava and 25% of total production (Thresh et mosaic virus (EACMV). A further al., 1997). geminivirus species has been found since in southern Africa and named In 1983, the first sequence of a South African cassava mosaic virus geminivirus to be determined and (SACMV) (Berrie et al., 1998). published was that of African cassava mosaic virus (ACMV) (Stanley and Gay, During the 1990s, an unusually 1983). Later, this virus was included in severe and damaging form of CMD the genus Begomovirus of the family spread rapidly through Uganda and the Geminiviridae (Briddon and Markham, wider East African region and came to 1995), consisting of viruses with be referred to as the CMD “pandemic” bipartite genomes transmitted by the (Otim-Nape et al., 1997; Legg, 1999). A novel recombinant geminivirus was * John Innes Centre (JIC), Norwich Research shown to be consistently associated Park, Colney, Norwich, UK. with the pandemic (Zhou et al., 1997), ** International Institute of Tropical Agriculture-Eastern and Southern Africa commonly in dual infections with Regional Center (IITA-ESARC), Kampala, ACMV that gave rise to the severest Uganda. symptoms (Harrison et al., 1997). The *** IITA-ESARC, Kampala, Uganda, and Natural Resources Institute, University of so-called Uganda variant has been Greenwich, Chatham Maritime, Kent, UK. shown to be a recombinant between

77 Whiteflies and Whitefly-borne Viruses in the Tropics

EACMV and ACMV, with ACMV The PCR procedure and the primers providing an approximately 400 bp utilized have been described previously fragment of the coat protein gene; this (Briddon and Markham, 1994). Typically virus is therefore best described as 1 µL, or 1 µL of a 10-fold dilution, of the nzania EACMV-Uganda (EACMV-Ug). DNA extract was used in the Subsequent studies have identified DNA amplification reactions. Reaction volumes B components for EACMV-Ug, have of 100 µL were used and amplification confirmed the occurrence in Uganda of products were cleaned further by phenol EACMV and have demonstrated extraction and ethanol precipitation infectivity through the generation of before use in the restriction fragment infectious clones (Pita et al., 2001). length polymorphism (RFLP) analysis. ozambique At the beginning of this project it Typically 10 µL (of the original was apparent that the different 100 µL reaction volume) of the PCR begomovirus species and strains product was used for restriction enzyme associated with CMD did not occur digestion in a 100 µL reaction volume necessarily in different regions of Africa. following the manufacturers’ (Gibco-BRL) It was therefore a major objective of the instructions. After a 4-hour digestion, work to survey the diversity of cassava RFLP products were resolved on 1% mosaic geminiviruses in the main agarose gels in TBE buffer stained with cassava growing areas of Africa and ethidium bromide. For most cases this determine their geographical sufficed to determine the restriction distribution. pattern. However, for a few samples the amplification was poor. This necessitated blotting of the gel to Hybond membranes Methods and Materials (Amersham) and probing with radioactively labeled probes. As a rule, all Cassava is normally propagated by gels were blotted to Hybond for stem cuttings and national collaborators subsequent analysis. therefore collected stem samples from CMD-infected plants during their Restriction patterns with particular project surveys, and shipped these to viruses were predicted from published the John Innes Centre (JIC), UK. The sequences (ACMV: Stanley and Gay, material was planted in compost in 1983; Morris et al., 1990. EACMV: Hong containers (7.6 cm diameter) and et al., 1993. EACMV-Ug: Zhou et al., maintained at about 25 °C (± 5 °C) in 1997). glasshouses, with supplementary lighting between October and April. For each of the virus species encountered in the restriction mapping Samples of nucleic acids for procedures, and any viruses with polymerase chain reaction (PCR) restriction maps that did not conform to analysis were extracted from 0.1 g known patterns, full length DNA A samples of infected leaf tissue using the genomic components were cloned for Nucleon Phytopure plant DNA sequence analysis. Clones were produced extraction kit (Amersham) essentially as by PCR amplification with specific described by the manufacturer. The abutting, non-overlapping primers main modification of the procedure was (Briddon et al., 1993). The sequences of that liquid nitrogen was used in place these primers were obtained by cloning of dry ice. Samples were stored at the PCR products produced with –200 °C and used directly in the PCR universal primers into pGem T-Easy reactions. vectors (Promega). Sequences were

78 Diversity of African Cassava Mosaic Disease determined by dideoxynucleotide chain Results termination sequencing using the PCR- based BIG DYE kit (Perkin Elmer Table 1 summarizes the analysis of the Cetus) and specific internal primers samples. About 13% of samples re- (Gibco-BRL). Reaction products were grew initially without symptoms, resolved on an ABI 377 automated although 14% of these (2% of the total) sequencer. Sequence information was did eventually show symptoms after stored, assembled and analysed using more than 3 months. Less than 1% Version 7 of the program library of the recovered after initially showing Genetics Computer Group. symptoms. Table 2 shows the viruses

Table 1. Summary of analysis of cassava cuttings collected during field surveys. Sampled plants were all showing symptoms when collected.a

Country Total no. Dead Re-grown Re-grown Healthy > Infected > DNA PCR of stems healthy infected infected healthy extract DNA-A

Tanzania 80 9 0 71 0 0 39 26 Kenya 45 15 0 30 0 2 14 5 Uganda 126 28 11 87 0 0 65 24 Zimbabwe 1 0 0 1 0 0 1 1 Malawi 23 19 2 2 0 0 Madagascar 193 1 39 153 1 2 30 18 Benin 96 46 7 43 5 2 35 15 Nigeria 68 21 27 20 7 0 25 12 Ghana 4 0 0 4 0 0 4 0 Cameroon 32 24 4 4 0 0 4 0 Totalb 671 163 (24) 90 (13) 418 (62) 13 (14) 6 (1) 217 (51) 101 (47) a. “Healthy” indicates plants showing no disease symptoms after about 3 months of re-growth. “Healthy > infected” indicates plants that initially showed no symptoms on re-growth but developed disease later; “Infected > healthy” indicates the converse. Other data: not sampled 39 (9%); yet to be identified 68 (16%). b. ( ) Percentage of totals; for DNA extract ( ) percentage of successful viral DNA extractions made from cassava mosaic disease samples; for PCR DNA-A ( ) percentage of DNA A products successfully amplified from viral DNA.

Table 2. Summary of virus species isolated following restriction mapping using Mlu1, Dra1 and EcoR1 on polymerase chain reaction products following the use of universal primers to amplify Begomovirus genome component A.

Country Virus speciesa Total

ACMV EACMV EACMV-Ug SACMV Dual

Tanzania 3 5 1 8 Kenya 1 3 1 4 Uganda 6 2 15 3 23 Zimbabwe 1 1 Madagascar 12 4 16 Benin 12 3 15 Nigeria 8 8 Totalsb 42 (56) 17 (23) 15 (20) 1 [5] 75 a. ACMV, African cassava mosaic virus; EACMV, East African cassava mosaic virus; EACMV-Ug, EACMV-Uganda; SACMV, South African cassava mosaic virus; Dual, combinations. b. ( ) percentage of total; [ ] included in total as single infections.

79 Whiteflies and Whitefly-borne Viruses in the Tropics identified from the samples analysed: Discussion ACMV comprised 56%, EACMV comprised 43% (of which 53% were The results obtained during this EACMV and 47% EACMV-Ug) and 7% screening project indicate that a PCR/ were dual infections (of which all RFLP-based approach is able except one were combinations of ACMV consistently to identify the viruses and EACMV-Ug). Samples from West associated with CMD and identify Africa comprised less than one-third of species or strains that differ the total identified (31%) but of these significantly from those previously only 13% were EACMV, while this virus reported. It is unclear whether a accounted for 56% of East Africa protocol based solely upon PCR samples. amplification with diagnostic primers for each virus species/strain could Full-length clones of the DNA A achieve the same. components of CMD-associated begomoviruses were obtained from It is surprising, based on experience isolates originating from Uganda and of related viruses, to find the genetic from Zimbabwe. The sequence of the stability which is evident from the very clone originating from Uganda shows low variability detected both in the this to be highly similar to the sequence analysis of clones and from sequence of EACMV-Ug reported by the RFLP analysis. However, this Zhou et al. (1997). The DNA A stability makes RFLP analysis a component of this virus is essentially particularly useful diagnostic tool for EACMV with an approximately 450 bp these viruses. This method also enables insertion of an ACMV sequence in the plants that are dual-infected to be centre of the coat protein gene. The correctly diagnosed because the overlaid restriction pattern of the virus RFLP patterns are evident. Possibly the originating from Zimbabwe was entirely major drawback to the technique is that novel, with no counterpart in the high-quality samples of nucleic acids published literature. Sequence analysis are essential but these may be difficult of this clone showed it to have most to extract from cassava. This was most similarity to SACMV (Berrie et al., evident in the case of the dried cassava 1998). The sequence published by leaf samples sent to JIC from Africa, Berrie et al. (1998) covers only the coat from which no successful PCR protein gene of SACMV to which the amplification was achieved. The reason sequence of the Zimbabwe clone shows for this failure remains unclear. It is 86% similarity. In the absence of the possible that sample degradation occurs full sequence of SACMV we must under tropical (hot and humid) provisionally conclude that the conditions; fungal growth was Zimbabwe virus is distinct from, but particularly evident on some of the leaf most closely related to, SACMV. Efforts samples. This problem has been are now under way to obtain the DNA encountered previously where even B genomic components of these two samples dried under optimal conditions viruses for infectivity studies. Attempts were not suitable for DNA extraction also are continuing to produce full- and subsequent begomovirus PCR length clones of EACMV and ACMV amplification. However, the cassava leaf originating from Madagascar. Although samples collected from the glasshouse amplification of these viruses has been and dried immediately yielded achieved, it has not yet proven possible amplification products more than 1 year to clone them. later, after storage at room temperature but in a dry atmosphere.

80 Diversity of African Cassava Mosaic Disease

It is clear from the recent reports of determination guaranteed its initiation. new strains/species of begomoviruses We also wish to thank all members of identified in association with CMD that the project team who sent material the diversity of these agents is greater during their regional surveys; without than initially expected. The procedures their help and endeavours, the project utilized for the analysis of the diversity would not have been possible. We also of begomoviruses infecting cassava thank Dr. Richard Gibson and Mr. Ian across Africa were devised to be all Robertson for sending extra samples encompassing. The use of so-called for inclusion. The work was also “universal primers”, designed with supported by the John Innes Centre, degeneracy so as to amplify all whitefly- which is grant-aided by the transmitted begomoviruses, ensures Biotechnology and Biological Sciences that all possible begomoviruses in a Research Council. The viruses were particular sample are amplified. The imported and held under the authority use of specific primers in an analysis of the Ministry of Agriculture Fisheries like this would risk missing possible and Food, under the Plant Health new viruses, or mutants of viruses, (Great Britain) Order 1993, Licence which have been described previously. Numbers PHL11/2557(4/1998) and The universal primers used (Briddon PHL/2300(6/1998). and Markham, 1994) are well characterized and have been able thus far to amplify all begomoviruses against References which they have been tested. Berrie, L. C.; Palmer, K. E.; Rybicki, E. P.; The use of RFLP analysis on PCR- Rey, M. E. C. 1998. Molecular characterization of a distinct South amplified DNA allows several diagnostic African cassava infecting restriction enzymes to be used on a geminivirus. Arch. Virol. 143: single sample to identify a virus. The 2253-2260. initial identification of a virus species/ strain by a particular restriction Briddon, R. W.; Markham, P. G. 1994. pattern was subsequently confirmed by Universal primers for the PCR sequence analysis of a full-length amplification of dicot-infecting geminiviruses. Mol. Biotechnol. clone. Full characterization of a clone 1:202-205. requires infectivity studies on cassava (Briddon et al., 1998). Our work has Briddon, R. W.; Markham, P. G. 1995. shown that the use of alternate hosts, Geminiviridae. In: Murphy, F.A.; such as Nicotiana benthamiana Domin Fauquet, C. M.; Bishop, D. H. L.; (Solanaceae), can lead to selection of Ghabrial, S. A.; Jarvis, A. W.; mutants with atypical biological Martelli, G. P.; Mayo, M. A.; Summers, M. D. (eds.). Virus characters (Liu et al., 1997) and thus is taxonomy: Sixth report of the not appropriate for diagnostic studies. International Committee on Taxonomy of Viruses. Springer, Vienna, AT. p. 158-165. Acknowledgements Briddon, R. W.; Prescott, A. G.; Lunness, The authors wish to thank the Danish P.; Chamberlin, L. C. L.; Markham, International Development Agency for P. 1993. Rapid production of full- length, infectious geminivirus providing funds for this project, and all clones by abutting primer PCR the participants of the Systemwide (AbP-PCR). J. Virol. Methods 43: Program on Integrated Pest 7-20. Management whose enthusiasm and

81 Whiteflies and Whitefly-borne Viruses in the Tropics

Briddon, R. W.; Liu, S.; Pinner, M. S.; Morris, B.; Coates, L.; Lowe, S.; Markham, P. G. 1998. Infectivity of Richardson, K.; Eddy, P. 1990. African cassava mosaic virus clones Nucleotide sequence of the to cassava by biolistic inoculation. infectious cloned DNA components Arch. Virol. 143:1487-1492. of African cassava mosaic virus (Nigerian strain). Nucleic Acids Res. Calvert, L. A.; Thresh, J. M. 2002. The 18:197-198. viruses and virus diseases of cassava. In: Hillocks, R. J.; Thresh, Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; J. M.; Bellotti, A. C. (eds.). Cassava: Baguma, Y.; Ogwal, S.; Semakula, Biology, production and utilization. G. N.; Acola, G.; Byabakama, B.; CAB International, Wallingford, GB. Martin, A. 1997. Cassava mosaic p. 237-260. virus disease in Uganda: The current pandemic and approaches Harrison, B. D.; Robinson, D. J. 1988. to control. Natural Resources Molecular variation in vector-borne Institute (NRI), Chatham, GB. 65 p. plant viruses: Epidemiological significance. Philos. Trans. R. Soc., Pita, J. S.; Fondong, V. N.; Sangare, A.; Lond., B, Biol. Sci. 321:447-462. Otim-Nape, G. W.; Ogwal, S.; Fauquet, C. 2001. Recombination, Harrison, B. D.; Zhou, X.; Otim-Nape, G. pseudorecombination and W.; Liu, Y.; Robinson, D. J. 1997. synergism of geminiviruses are Role of a novel type of double determinant keys to the epidemic of infection in the geminivirus-induced severe cassava mosaic disease in epidemic of severe cassava mosaic Uganda. J. Gen. Virol. 82:655-665. in Uganda. Ann. Appl. Biol. 131:437-448. Stanley, J.; Gay, M. G. 1983. Nucleotide sequence of cassava latent virus Hong, Y. G.; Robinson, D. J.; Harrison, B. DNA. Nature 301:260-262. D. 1993. Nucleotide sequence evidence for the occurrence of three Thresh, J. M.; Otim-Nape, G. W.; Legg, J. distinct whitefly-transmitted P.; Fargette, D. 1997. African geminiviruses in cassava. J. Gen. cassava mosaic disease: The Virol. 74:2437-2443. magnitude of the problem? Afr. J. Root Tuber Crops 2:13-19. Legg, J. P. 1999. Emergence, spread and strategies for controlling the Warburg, O. 1894. Die Kulturpflanzen pandemic of cassava mosaic virus Usambaras. Mitt. Dtsch. Schutz. disease in east and central Africa. 7:131. Crop Prot. 18:627-637. Zhou, X.; Liu, Y.; Calvert, L.; Muñoz, C.; Liu, S.; Bedford, I. D.; Briddon, R. W.; Otim-Nape, G. W.; Robinson, D. J.; Markham, P. 1997. Efficient Harrison, B. D. 1997. Evidence that whitefly transmission of African DNA-A of a geminivirus associated cassava mosaic geminivirus with severe cassava mosaic disease requires sequences from both in Uganda has arisen by genomic components. J. Gen. Virol. interspecific recombination. J. Gen. 78:79-94. Virol. 78:2101-2111.

82 Sweetpotato Virus Disease in East Africa

CHAPTER 1.12 Serological Analysis of Sweetpotatoes Affected by Sweetpotato Virus Disease in East Africa

Valente Aritua*, Richard Gibson** and Josef Vetten***

Introduction et al., 1998). SPFMV has a worldwide distribution, is readily spread by Infection of sweetpotato (Ipomoea aphids such as Myzus persicae (Sulzer) batatas [L.] Lam.) by viruses is a major in a non-persistent manner but cause of yield reduction worldwide. At probably is not spread by seed least 13 viruses are reported to infect transmission (Cadena-Hinojosa et al., sweetpotato naturally (Moyer and 1981a). Various strains of SPFMV have Larsen, 1991). Most of them are insect- been differentiated, mostly by transmitted, mainly by whitefly or symptoms produced on certain aphid species. In East Africa, sweetpotato cultivars. In USA, two symptoms of sweetpotato virus disease strains are recognized—the russet (SPVD) were first reported on crack strain or internal cork strain, sweetpotato in 1945 (Hansford, 1945). which causes internal necrosis of the However, the presence of viruses was storage roots of some cultivars, and the not demonstrated until 1957 when common strain, which causes no such Sheffield (1957) associated two viruses symptoms. In East Africa, SPFMV- with sweetpotato plants having virus- infected sweetpotato plants typically like symptoms, virus A being aphid- exhibit no symptoms on either roots or transmitted and virus B being whitefly- foliage. SPMMV is considered to be transmitted. Virus A was later whitefly-borne (Hollings and Stone, identified as Sweetpotato feathery mottle 1976). virus (SPFMV) but the identity of the whitefly-transmitted virus remained The whitefly-transmitted unclear. Subsequent efforts to component virus of SPVD has been determine the range of viruses named as Sweetpotato chlorotic stunt occurring in East African sweetpotato virus (SPCSV) largely on the basis of crops confirmed SPFMV as the most the symptoms of West African isolates frequently found virus but also expressed in the indicator plant revealed the presence of Sweetpotato Ipomoea setosa Ker Gawl. (Schaefers mild mottle virus (SPMMV), Sweetpotato and Terry, 1976). SPCSV is latent virus (SPLV) and Sweetpotato synonymous (Gibson et al., 1998) with chlorotic fleck virus (SPCFV) (Carey Sweetpotato sunken vein virus (Cohen et al., 1992) and SPVD-associated * Makerere University, Kampala, Uganda. closterovirus. An East African strain of ** Natural Resources Institute (NRI), SPCSV (SPCSV-SEA) was identified University of Greenwich, Chatham using monoclonal antibodies (MABs) Maritime, Kent, UK. *** Biologische Bundesanstalt für Land und prepared against an Israeli isolate of Forstwirtschaft, Braunschweig, Germany. SPCSV as well as against bacterially

83 Whiteflies and Whitefly-borne Viruses in the Tropics expressed coat protein of a Kenyan Collection and Analysis of isolate (Hoyer et al., 1996; Gibson et Samples al., 1998). Subsequent analyses using enzyme-linked immunosorbent assays Project partners carried out field (ELISA) led to the description of two surveys according to a common distinct serotypes: SEA1 as originally protocol. Additional information on the described from Kenya, and SEA2 (Alicai timing and coverage of the surveys are et al., 1999). In East Africa, SPCSV given in the respective country papers causes only moderate stunting and in the present volume (Chapters 1.6, yellowing or purpling of the lower and 1.7 and 1.8). The authors analysed middle leaves when it occurs alone samples collected by project partners (Gibson et al., 1998). Co-infection of as follows: those from Uganda were SPFMV and SPCSV is very common tested fresh at Namulonge Research and the severe symptoms of SPVD Institute; dried samples from Kenya result from a synergistic interaction and Tanzania were analysed at the between SPCSV and SPFMV. SPVD is Biologische Bundesanstalt für Land the most important disease of und Forstwirtschaft (BBA), sweetpotato in Africa (Geddes, 1990). Braunschweig, Germany. The samples Depending on the genotype, SPVD were analysed using polyclonal antisera symptoms comprise leaf strapping, or monoclonal antibodies to SPCSV, mottling, vein clearing, puckering and SPFMV and SPMMV using triple stunting of sweetpotato plants antibody sandwich (TAS), double (Schaefers and Terry, 1976). antibody sandwich (DAS), or nitrocellulose membrane (NCM) ELISA. The respective distributions of SPMMV, SPCSV and its synergistic More than 200 sweetpotato leaf partner, SPFMV, are still poorly samples were collected from plants understood in the major sweetpotato showing possible symptoms of viral growing regions of East Africa, infection. Field symptoms observed although it has long been known that consisted of leaf strapping, yellowing, these viruses can cause large yield purpling, mottling, vein clearing and losses (Mukiibi, 1977). The growing stunting of sweetpotato plants. Since importance of sweetpotato for food incidence varied among and within the security in Africa has increased three countries, different numbers of international exchange of sweetpotato samples were collected from each germplasm and thus increased the country and target area. need to identify and establish the distribution of the common viruses occurring on the crop. Such Reactions of the Diseased information is an important Samples prerequisite for the development and appropriate release of cultivars Table 1 shows the numbers of samples resistant to the prevailing viruses/ by country and target area that tested diseases. positive for SPFMV, SPSCV-SEA and/ or SPMMV. Both SPFMV and SPCSV The following work investigates the occurred throughout the sampled distributions of SPMMV, SPCSV and its areas. Overall, SPFMV was most serotypes, and SPFMV in Uganda, frequently detected, in 151 out of the Kenya and Tanzania. 243 samples. It was most frequently

84 Sweetpotato Virus Disease in East Africa

Table 1. Results of enzyme-linked immunosorbent assay (ELISA) tests using antiserum/monoclonal antibodies to viruses of sweetpotato for sweetpotato leaf samples collected in Uganda, Kenya and Tanzania.

Country Survey area Number of positive samplesa

SPFMV 1 SPCSV (SEA) 2 SPMMV 3

Uganda Northern 12/38 14/38 0/36 Eastern 13/36 14/36 0/36 Central 26/38 34/38 2/38 Western 19/40 38/40 7/40 % frequency 46 66 6

Kenya Coast 2/4 1/4 0/4 Western 9/12 5/12 4/12 Nyanza 9/9 5/9 1/9 % frequency 80 44 20

Tanzania N. Coast 17/21 5/21 0/21 S. Coast 23/23 8/23 0/23 Lake zone 21/22 16/22 7/22 % frequency 92 44 11 a. Data are number of positive samples / number of samples tested. SPFMV 1, Sweetpotato feathery mottle virus: negative control, healthy Ipomoea setosa; positive control, SPCSV-Ky38 in I. setosa. SPCSV (SEA) 2, East African strain of Sweetpotato chlorotic stunt virus: negative control, healthy sweetpotato; positive control, SPFMV-46b in Nicotiana tabacum L. SPMMV 3, Sweetpotato mild mottle virus: negative control, healthy sweetpotato; positive control, SPMMV-DDR (German Democratic Republic). Values are considered as positive where the mean readings were at least two times greater than the mean of the healthy controls. detected in samples from Tanzania Some samples did not react (92%), slightly less frequently (80%) in positively to any of the viruses tested, those from Kenya and least frequently even though they appeared to show (46%) in samples from Uganda. One virus symptoms in the field. Low virus hundred and forty samples reacted concentration and the irregular positively to MABs specific to the East distribution of SPFMV in sweetpotato African strain of SPCSV. Frequencies of plants are frequently cited as obstacles detection were 66% in samples from to the reliable use of ELISA for virus Uganda, 44% in samples from detection in sweetpotato (Cadena- Tanzania and 44% from Kenyan Hinojosa and Campbell, 1981b). Other samples. SPMMV was detected only in possible explanations are that the samples collected from areas around presence of high concentrations of Lake Victoria in Uganda (6% phenols, latex or other inhibitors incidence), Kenya (20%) and Tanzania adversely affected the reagents used in (11%) (Table 1). Co-infections of SPFMV these tests and that tests on samples + SPCSV were more frequent (32% of from Tanzania and Kenya were done samples) than the occurrence of either on dried leaf samples stored for SPFMV (22%) or SPCSV (16%) alone. various lengths of time. Moreover, This association was expected because symptoms of viral infection can be the two viruses co-exist synergistically difficult to recognize under certain (Schaefers and Terry, 1976). All circumstances and may have led to the combinations of viruses occurring collection of some uninfected samples, together in samples also were detected particularly in locations where the but more rarely (Table 2). viruses occur less frequently, as was

85 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 2. Single and multiple occurrences of viruses of sweetpotato in leaf samples from three East African countries.

Country Nothing Sweetpotato virusesa detected SPFMV SPCSV SPMMV SPFMV + SPFMV + SPCSV + SPFMV + alone alone alone SPCSV SPMMV SPMMV SPCSV + SPMMV

Uganda 36 15 40 1 54 0 1 5 Tanzania 0 30 0 0 29 1 0 6 Kenya 2 9 0 1 9 2 1 1 Total 38 54 40 2 92 3 2 12 % 1622161381 1 5 a. SPFMV, Sweetpotato feathery mottle virus; SPCSV, Sweetpotato chlorotic stunt virus and SPMMV, Sweetpotato mild mottle virus. the case in northern and eastern both serotypes reacting with one regions of Uganda. Alternatively, since (MAB 2G8) but only SEA2 reacting with all samples tested were selected on the the other (MAB 6D12) (Alicai et al., basis of apparent virus symptoms, the 1999). Serotype SEA2 was detected in lower rates of detection of virus in 66 of the 96 SPCSV-positive samples, samples from more easterly areas may while SEA1 was detected in 30 of the indicate the presence there of viruses SPCSV positive samples. Serotype other than SPCSV, SPFMV or SPMMV SEA2 was found in all districts except (i.e., viruses for which no tests were for Tororo, Masindi and Iganga, carried out). although it was found most frequently in the central and western survey In Uganda, the presence of two areas. Serotype SEA1 was distributed serotypes of SPCSV was confirmed more evenly across the country and, in using two MABs (Hoyer et al., 1996), contrast to the SEA2, occurred more

Table 3. Distribution of two Sweetpotato chlorotic stunt virus (SPCSV) serotypes in sweetpotato leaf samples from 12 districts of Uganda.

Target area District No. positive samples SPCSV serotype

MAB 2G8 MAB 6D12 SEA2 SEA1

Northern Apac 4 1 3 1 Lira 4 1 3 1 masindi 5 5 0 5 Total 6 7

Eastern Tororo 1 1 0 1 Palisa 8 6 2 6 Iganga 5 5 0 5 Total 2 12

Central Luwero 11 0 11 0 Mpigi 9 2 7 2 Mukono 11 4 7 4 Total 25 6

Western Masaka 13 3 10 3 Rakai 12 0 12 0 Mubende 13 2 11 2 Total 33 5

86 Sweetpotato Virus Disease in East Africa

frequently in the northern and eastern References survey areas (Table 3). These results confirm the heterogeneity of SPCSV, at Alicai, T.; Fenby, N. S.; Gibson, R. W.; least at the serotype level. Further Adipala, E.; Vetten, H. J.; Foster, G. D.; Seal, S. E. 1999. Occurrence of studies would therefore be useful to two serotypes of Sweetpotato investigate the diversity of this group of chlorotic stunt virus in East Africa viruses and clarify any implications for and their associated differences in disease management. coat protein and HSP70 homologue gene sequences. Plant Pathol. 48:718-726.

Conclusions Aritua, V.; Adipala, E.; Carey, E. E.; Gibson, R. W. 1998. The incidence SPCSV is a common whitefly-borne of sweetpotato virus disease and virus in sweetpotato throughout East virus resistance of sweetpotato Africa, often occurring together with grown in Uganda. Ann. Appl. Biol. SPFMV, with which it acts 132:399-411. synergistically to produce the severe disease SPVD. In contrast, SPMMV is Cadena-Hinojosa, M. A.; Campbell, R. N. 1981a. Characterisation of four largely restricted to the Lake Victoria aphid transmitted sweetpotato region of East Africa. viruses. Phytopathology 71: 1086-1089. Differences in the distribution of two SPCSV serotypes have been Cadena-Hinojosa, M. A.; Campbell, R. N. confirmed. Both SEA1 and SEA2 are 1981b. Serological detection of widely distributed throughout Uganda feathery mottle virus strains in sweetpotatoes and Ipomoea but SEA1 is more prevalent in northern incarnata. Plant Dis. 65:412-414. and eastern Uganda and SEA2 is more prevalent in central and western Carey, E. E.; Mwanga, R. O. M.; Fuentes, Uganda. These areas correspond with S.; Kasule, S.; Macharia, C.; areas of lesser and greater whitefly Gichuki, S. T.; Gibson, R. W. 1998. abundance and slower and faster rates Sweetpotato viruses in Uganda and of spread (Aritua et al., 1998) but Kenya: Results of a survey. In: Procs. Sixth Triennial Symposium whether this relationship is causal is of the International Society of unclear. Tropical Root Crops-Africa Branch (ISTRC-AB), 22-28 October 1995, Lilongwe, MW. p. 457-461. Acknowledgements Cohen, J.; Franck, A.; Vetten, H. J.; The technical assistance of Mr. Titus Lesemann, D. E.; Loebenstein, G. Alicai during sample analysis is 1992. Purification and properties of closterovirus-like particles gratefully acknowledged. The Crop associated with a whitefly- Protection Programme of the UK transmitted disease of sweetpotato. Department for International Ann. Appl. Biol. 121:257-268. Development funded Dr. Richard Gibson’s work. Geddes, A. M. 1990. The relative importance of crop pests in sub- Saharan Africa. Bulletin no. 36, Natural Resources Institute (NRI), Chatham, GB.

87 Whiteflies and Whitefly-borne Viruses in the Tropics

Gibson, R. W.; Mpembe, I.; Alicai, T.; Moyer, J. W.; Larsen, R. C. 1991. Carey, E. E.; Mwanga, R. O. M.; Management of insect vectors of Seal, S. E.; Vetten, H. J. 1998. viruses infecting sweetpotato. In: Symptoms, aetiology and serological Jansson, R. K.; Raman, K. V. (eds.). analysis of sweetpotato virus Sweetpotato pest management: A disease in Uganda. Plant Pathol. global perspective. Westview Press, 47:95-102. Denver, Colorado, USA. p. 341-358.

Hansford, C. G. 1945. Ugandan plant Mukiibi, J. 1977. Effects of mosaic virus diseases. Part IV. Diseases of on the yields of sweetpotato in sweetpotato. East Afr. Agric. J. Uganda. Procs. Fourth International 10:126-127. Symposium on Tropical Root Crops. International Development Hollings, M.; Stone, O. M. 1976. Research Centre (IDRC), Ottawa, Purification and properties of CA. p. 169-170. sweetpotato mild mottle, a whitefly- borne virus from sweetpotato Schaefers, G. A.; Terry, E. R. 1976. Insect (Ipomoea batatas) in East Africa. transmission of sweetpotato Ann. Appl. Biol. 82:511-528. diseases in Nigeria. Plant Pathol. 66:642-645. Hoyer, U.; Maiss, E.; Jekmann, W.; Lessemann, D. E.; Vetten, H. J. Sheffield, F. M.L. 1957. Virus diseases of 1996. Identification of coat protein sweetpotato in East Africa. 1. gene of sweetpotato sunken vein Identification of viruses and their closterovirus isolate from Kenya insect vectors. Phytopathology and evidence for serological 47:582-587. relationship among geographically diverse closterovirus isolates from sweetpotato. Phytopathology 86:744-750.

88 Factors Associated with Damage

CHAPTER 1.13 Factors Associated with Damage to Sweetpotato Crops by Sweetpotato Virus Disease

Richard Gibson*, Valente Aritua** and Simon Jeremiah***

Introduction are infected with SPCSV quickly develop SPVD, either through an Sweetpotato virus disease (SPVD) is the activation of a latent SPFMV infection most destructive disease of sweetpotato or through increased susceptibility to (Ipomoea batatas [L]) in Africa (Geddes, aphid-borne spread of the virus. In this 1990). It is caused by a combined manner, infection with SPCSV is the infection of the aphid-borne trigger for SPVD and damage by the Sweetpotato feathery mottle virus disease has long been linked with high (SPFMV) and the whitefly-borne populations of its vector, the whitefly Sweetpotato chlorotic stunt virus Bemisia tabaci (Gennadius) (Sheffield, (SPCSV) (Schaefers and Terry, 1976; 1957). Gibson et al., 1998b). Infection with SPFMV alone causes no obvious SPCSV is transmitted semi- symptoms in African sweetpotato persistently by B. tabaci; feeds lasting a varieties (Gibson et al., 1997), while few hours are required for both efficient infection with SPCSV alone causes only acquisition and inoculation (Cohen et moderate stunting, yellowing or al., 1992). B. tabaci that colonize purpling of middle and lower leaves sweetpotato in Africa cannot readily and some yield loss (Gibson et al., colonize cassava (Manihot esculenta 1998b). By contrast, SPVD causes Crantz) and vice-versa (Burban et al., severe plant stunting and small, 1992; Legg, 1996). As a result, there distorted leaves with either chlorotic may be no direct link between whitefly mosaic or vein clearing (Schaefers and numbers on sweetpotato and whitefly Terry, 1976). The yield of affected numbers on cassava, or between plants generally is reduced by more transmission of SPCSV and of African than 50% (Mukiibi, 1977; Hahn, 1979; cassava mosaic virus, even where Ngeve and Boukamp, 1991). It appears sweetpotato and cassava are that SPCSV has a synergistic effect on intercropped. Thus, the epidemic of SPFMV (Rossel and Thottappilly, 1987; cassava mosaic disease and B. tabaci in Aritua et al., 1998a): field plants that cassava crops in East Africa seems to have had little effect on the incidence of SPVD. This is particularly fortunate * Natural Resources Institute (NRI), because sweetpotato often provides the University of Greenwich, Chatham Maritime, Kent, UK. most readily available alternative crop ** Makerere University, Kampala, Uganda. to cassava. *** Root and Tubers Programme of the Lake Zone Agricultural Research and Development Institute (LZARDI), Ukiriguru, Sweetpotato is an important crop Mwanza, Tanzania. for subsistence farmers throughout

89 Whiteflies and Whitefly-borne Viruses in the Tropics

Uganda, generally being grown Methods and Results continuously throughout the year. Planting occurs whenever soil moisture Monthly surveys of whiteflies and is adequate and so is concentrated SPVD were carried out at 10 farmers’ during, although not exclusively limited fields around each of six towns in to, the rains, which in most areas of Uganda (Soroti, Busia, Iganga, Uganda occur from March to June and Namulonge, Kanoni and Masaka), from from September to November. Crops January 1998 to April 1999. All the are usually maintained for at least sites are at about the same altitude 6 months, cropping frequently being and lie roughly along a south-east/ extended for up to 1 year by piecemeal north-west axis. Sweetpotato fields, harvesting, using a stick to tease out 4 to 8 months old, were chosen while individual mature tubers (Bashaasha et travelling along a rural road or path al., 1995). around each town, stopping at intervals of about 1 km or at the first SPVD has been reported to be field observed thereafter, different prevalent at moderate altitudes, 50 to plantings being sampled each month. 600 m above sea level, in East Africa Whitefly numbers in each field were (Sheffield, 1953). However, in Uganda, assessed by counting the number of it was rare in the eastern districts of whitefly adults observed while turning Soroti, Tororo and Busia but reached over leaves for a 1-minute period. Ten damaging levels of incidence in central different parts of each crop were Mpigi District, as well as in the sampled in this manner. A more southern districts of Masaka and detailed account of the sampling Rakai, even though all locations were method is provided in Aritua et al. at a similar mid-altitude (Aritua et al., (1998b). SPVD was assessed by 1998b). The main aim of this project counting the proportion of affected therefore was to understand the plants among 30 plants selected at underlying causes of the different random along a V-shaped track incidences, so as to provide a sound through each field. base from which to develop management practices. Recent work Whiteflies were relatively few at (Gibson et al., 1997; Aritua et al., Soroti and Busia except in February 1998b) identified three factors as being 1998 (Figure 1) and, in accord with associated with low SPVD incidence: this, SPVD was also rare there (Figure 2). Similarly, there were slightly (1) Whiteflies were rare on sweetpotato more of both whiteflies and SPVD at in some localities (for example, the Iganga. Although farmers’ fields at eastern districts of Uganda); Namulonge, Kanoni and Masaka all had relatively large and similar (2) The predominant sweetpotato numbers of whiteflies, SPVD at variety is resistant; and Namulonge averaged only 2% and at Masaka, 6%, reaching 18% at Kanoni. (3) Farmers select planting material Whitefly numbers were relatively high from symptomless parents. at all sites from January to April (inclusive). At those sites where SPVD Our work therefore focused on was common, its incidence remained links between SPVD incidence and (a) relatively low until February, consistent whitefly numbers, (b) varietal with there being a latent period of resistance and (c) farmers’ plant health about 1 month (Gibson et al., 1998a) management practices. between infection by the whiteflies and

90 Factors Associated with Damage

10 Mean whiteflies per minute 5

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Date of observation

Soroti Busia Iganga Namulonge Masaka Kanoni

Figure 1. Mean whitefly numbers in six districts of Uganda (January 1998-April 1999). Each data point represents the mean for 10 fields.

Mean SPVD incidence (%) 5

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Date of observation

Soroti Busia Iganga Namulonge Masaka Kanoni

Figure 2. Sweetpotato virus disease (SPVD) incidence in farmers’ fields in six districts of Uganda (January 1998-April 1999). appearance of SPVD. The incidence of From the point of view of SPVD then remained relatively high understanding the relationship until August, partly perhaps as a result between vector dynamics and disease of continuing infection but probably incidence over time, sampling farmers’ also because of the survival of affected fields had the disadvantages that they plants. were not necessarily planted with the

91 Whiteflies and Whitefly-borne Viruses in the Tropics same cultivars and that the planting (4) There was no correlation between material already might have been whitefly numbers and final SPVD infected and, to a varying degree, with incidence (Table 1). SPVD. In order to eliminate these variables, three on-farm trials were An index of the level of local planted at each of the three most inoculum for each site at Namulonge contrasting sites, Soroti, Namulonge and Kanoni was also calculated from and Kanoni, using planting material “area of each field of sweetpotato within from Namulonge Agricultural and 100 m of each trial” × “incidence of Animal Production Research Institute SPVD in it”¸ “distance to the trial plot”. (NAARI) farm, with one series of Results are tabulated (Table 1) for the plantings in the second rains of 1997 first series of trials in which there was (Aritua et al., 1999) and another in the more virus spread. The value for local second rains of 1998. Whitefly inoculum correlated closely with SPVD numbers and SPVD incidence were incidence at both sites, suggesting that monitored monthly, from January to the greater spread of SPVD at Kanoni June, using the same procedures as in was due to the greater incidence of the surveys of farmers’ fields. In both SPVD in nearby farmers’ fields. series, the results were very similar to Examination of records of farmers’ those obtained by monitoring farmers’ crops in the two localities also revealed fields: that a highly SPVD-resistant variety called New Kawogo predominated at (1) No SPVD and few whiteflies (results Namulonge and the surrounding not shown) were found at Soroti; villages and this may explain why SPVD was rare in the farmers’ fields (2) Whiteflies were abundant at both there. At Kanoni, more susceptible Namulonge and Kanoni (Table 1); varieties predominated in farmers’ fields (Aritua et al., 1999). (3) But SPVD increased in incidence rapidly only at Kanoni (Figure 3); Links between SPVD incidence, and crop data and farmers’ management

Table 1. Relationship between sweetpotato virus disease (SPVD) incidence and (1) whitefly counts, (2) local inoculum and (3) whitefly numbers × local inoculum.

Farm SPVD Whitefliesb Local Whiteflies × incidencea inoculumc inoculum

Namulonge A 4.0 11.4 0 0 Namulonge B 1.5 13.2 100 1320.0 Namulonge C 0 14.0 120 1680.0 Kanoni D 18.5 10.5 1043 10951.5 Kanoni E 36.0 13.4 2298 30793.2 Kanoni F 23.5 9.4 1519 14278.6 Correlationd - -0.284 0.990 0.969 P - N.S. < 0.001 < 0.001 a. SPVD incidence (%): final assessment (June) in each on-farm trial. b. Whiteflies: mean monthly (January-June) whitefly counts from each on-farm trial. c. Local inoculum: index calculated from SPVD incidence × area of field ÷ distance from field for fields within 100-m radius of trial plots. d. Correlation of SPVD incidence with data in corresponding column.

92 Factors Associated with Damage

35 Experiment 1 (November 1997-April 1998) 30

20 % SPVD 15

35 Experiment 2 (October 1998-March 1999) 30

20 % SPVD 15

0 12 3 4 5 6 Months after planing

Nam-A Nam-B Nam-C Kan-D Kan-E Kan-F

Figure 3. Progress of sweet potato virus disease (SPVD) in initially disease-free sweetpotato planted during the second rains of 1997 (Experiment 1) and the second rains of 1998 (Experiment 2) at farms at Namulonge (Nam-A, Nam-B, Nam-C) and Kanoni (Kan-D, Kan-E, Kan-F).

practices also were examined by farm Most farmers, particularly those in surveys and farmer interviews in 1998- Tanzania not exposed as yet to the 99 (for methods, see Gibson et al., bitter lessons of the cassava mosaic 2000), focusing particularly on epidemic (which began in Uganda), did localities where SPVD was unusually not realize that SPVD involved a virus prevalent, notably Bukoba and spread by whiteflies, many considering Karagwe sub-districts of Kagera it to be caused by too much sun. Even District in Tanzania, and Rukungiri so, most farmers attempted to control and Mpigi Districts in Uganda (Table 2). SPVD, predominantly through the use

93 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 2. Importance to farmers of different measures that they already are using to manage sweetpotato virus disease (SPVD) (adapted from Gibson et al., 2000).

Locality Percentage of farmersa giving the rank indicated No. of farmers to a control measure

Symptomless Roguing Resistant Controlling Interviewed planting material cultivars SPVDb

1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd Tanzaniac Bukoba 100 0 0 0 33 0 0 0 0 15 (75) 20 Karagwe 80 0 0 20 10 0 0 0 0 10 (36) 28

Uganda Rukungiri 82 14 0 11 50 7 7 14 18 44 (88) 50 Mpigi 88 10 0 2 10 27 10 54 7 41 (82) 50 Soroti 100 0 0 0 17 17 0 17 0 6 (12) 50 Overall 87 9 0 7 28 13 6 25 9 116 (60) 198 a. Among those claiming to use specific measures to control SPVD. b. Number of farmers (in parentheses, percentage of farmers among those interviews) claiming to use specific control measures for SPVD. c. Farmers in other districts in Tanzania were not questioned on this subject. of plant health management tactics Conclusions (Table 2). In particular, most farmers carefully selected their planting Our results confirm that SPVD varies material from unaffected parent plants. considerably in incidence among sites A few also removed (rogued) diseased and that part of the variation in plants from young crops. However, incidence is associated with differences farmers often planted new crops close in the numbers of B. tabaci infesting to old diseased crops, even though the crop. Thus, the explanation for the incidence of SPVD and the distance rarity of SPVD in Soroti and Busia between crops were significantly seems to be that whiteflies are rare (P < 0.05) and negatively correlated. throughout much of the year (Figure 1) Farmers also appeared to make no but this still leaves unanswered the attempt to remove diseased plants from question of why fewer whiteflies should old and abandoned crops even when be present in Soroti and Busia than in they were close to newly planted fields. Mpigi District. Our investigation was This is consistent with farmers lacking too limited to identify positively the knowledge that SPVD is caused by an reason(s) for this. However, a major insect-transmitted virus. difference between Soroti and Busia Districts and Mpigi District is that the Although most farmers knew of first rains in Soroti and Busia Districts SPVD-resistant varieties, few ranked occur later and the second rains earlier resistance as a valuable management than in Mpigi District. This leads to a practice (Table 2), apparently because more prolonged dry season between the most resistant varieties had a poor second and the first rains in Soroti and and/or late yield. Despite this, they Busia, during which time most ranked new varieties with superior vegetation dries out, natural bushfires yield characteristics and resistance as are common and the resulting natural their top requirement. vegetation predominantly consists of low-growing bushes and grasslands. Rainfall being more evenly spread

94 Factors Associated with Damage throughout the year in Mpigi, (2) Roguing plants that develop vegetation continues to grow year- symptoms (used by a few farmers round and such fires rarely occur. As a early in the season); result, sweetpotato crops in this region often occur in locations sheltered by (3) Removal of crop debris (both roots tall trees and other vegetation. and foliage as both re-grow readily and are often diseased) from old High incidences of SPVD were fields (not currently used); associated with: (4) Separation of new crops from old (1) Large average numbers of diseased ones (not currently used); whiteflies found on the sweetpotato and crop during the year; (5) Increased use of resistant varieties (2) High peak whitefly populations, (favoured only in localities where which occurred during the hot dry acceptable resistant varieties were season (January-March); and available).

(3) Relatively large amounts of SPVD- Our work also suggests that it may affected plants nearby. be particularly important to apply plant health management tactics during the The importance of local inoculum hot dry season because this is when was shown directly by our on-farm field the whitefly vectors are most trials but was indicated also by survey abundant. However, it is important to data showing that SPVD incidence emphasize that, although they may increased as sweetpotato cropping seem obvious, none of these strategies intensity increased. A requirement for have been tested yet in field trials, to local inoculum is consistent with ensure that they are effective and SPCSV being only semi-persistently adoptable by farmers, and so cannot be transmitted and with B. tabaci being a officially recommended. Furthermore, relatively weak flier. Using pesticides to our farmer interviews indicated that reduce the population of the whitefly most farmers do not have a sound vector is highly unlikely to be a useful understanding of the cause and source tactic in SPVD management since, even of SPVD. Some of these control setting aside environmental risks, practices might therefore appear to human health concerns and the farmers to be inappropriate. tendency of whiteflies to develop Consequently, both participatory resistance to insecticides, the crop has research and training programs will be too low a value to justify the essential if plant health management investment in insecticides. Choice of strategies are to be developed and appropriate varieties is unlikely to be widely adopted. effective in reducing vector populations because earlier work has not identified Farmers in areas severely affected major differences in the numbers of by SPVD gave highest priority to the whiteflies on different sweetpotato need for superior, virus-resistant varieties (Aritua et al., 1998a). varieties (Table 3). Our work showed However, local inoculum could be that resistant varieties are valuable not reduced by: only because they suffer less SPVD damage but also because their (1) Planting disease-free cuttings widespread cultivation reduces (already used by most farmers); incidence of SPVD in remaining stands

95 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 3. Aspects of sweetpotato virus disease (SPVD) management on which farmers in severely affected localities would most like to receive assistance (adapted from Gibson et al., 2000).

Locality Percentage of farmersa giving a particular rank No. of farmers to the indicated management tactic

Superior, resistant Technical Chemical Requestingb Interviewed cultivars information control

1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd Bukoba 100 0 0 0 0 0 0 0 0 14 (70) 20 Rukungiri 60 21 0 23 6 2 17 15 0 48 (96) 50 Overall 69 16 0 18 5 2 13 11 0 62 (89) 70 a. Among those requesting specific measures to manage SPVD. b. Number of farmers (in parentheses, percentage among those interviewed) requesting assistance with management of SPVD. of susceptible varieties. Although Aritua, V.; Adipala, E.; Carey, E. E.; farmers also gave a low ranking to the Gibson, R. W. 1998b. The incidence use of their currently available of sweetpotato virus disease and virus resistance of sweetpotato resistant varieties (Table 2), this was grown in Uganda. Ann. Appl. Biol. because most farmers perceived their 132:399-411. resistant landraces to have major flaws. The higher yielding, resistant Aritua, V.; Legg, J. P.; Smit, N. E. J. M.; varieties such as those selected at Gibson, R. W. 1999. Effect of local NAARI in Uganda and at the inoculum on the spread of Agricultural Research Institute in sweetpotato virus disease: Widespread cultivation of a Ukiriguru, Tanzania, taking account of resistant sweetpotato cultivar may farmer preferences, should overcome be protecting susceptible cultivars. this obstacle. Plant Pathol. 48:655-661.

Bashaasha, B.; Mwanga, R. O. M.; Ocitti Acknowledgements p’Obwoya, C.; Ewell, P. T. 1995. Sweetpotato in the farming and food systems of Uganda: A farm survey We thank the many farmers who report. Centro Internacional de la patiently answered our questions and Papa (CIP), sub-Saharan Africa allowed their fields to be examined and Region, Nairobi, KE/National used for research purposes. We thank Agricultural Research Organisation the Director of NAARI and the Staff of (NARO), Kampala, UG. 63 p. the Ugandan Potato Programme for their support and the use of NAARI Burban, C.; Fishpool, L. D. C.; Fauquet, C.; Fargette, D.; Thouvenel, J. -C. facilities. We thank Mr. I. Mpembe, 1992. Host-associated biotypes Mr. R. P. Msabaha and Mr. J. within West African populations of Ndunguru for their collaborative the whitefly Bemisia tabaci inputs. (Genn.)(Hom., Aleyrodidae). J. Appl. Entomol. 113:416-423.

References Cohen, J.; Franck, A.; Vetten, H. J.; Lesemann, D. E.; Loebenstein, G. 1992. Purification and properties of Aritua, V.; Alicai, T.; Adipala, E.; Carey, closterovirus-like particles E. E.; Gibson, R. W. 1998a. Aspects associated with a whitefly- of resistance to sweetpotato virus transmitted disease of sweetpotato. disease in sweetpotato. Ann. Appl. Ann. Appl. Biol. 121:257-268. Biol. 132:387-398.

96 Factors Associated with Damage

Geddes, A. M. 1990. The relative Legg, J. P. 1996. Host-associated strains importance of crop pests in sub- within Ugandan populations of the Saharan Africa. Bulletin no. 36, whitefly Bemisia tabaci (Genn.) Natural Resources Institute (NRI), (Hom., Aleyrodidae). J. Appl. Chatham, GB. Entomol. 120:523-527.

Gibson, R. W.; Mwanga, R. O. M.; Kasule, Mukiibi, J. 1977. Effects of mosaic virus S.; Mpembe, I.; Carey, E. E. 1997. on the yields of sweetpotato in Apparent absence of viruses in Uganda. Procs. Fourth International most symptomless field-grown Symposium on Tropical Root Crops. sweetpotato in Uganda. Ann. Appl. International Development Biol. 130:481-490. Research Centre (IDRC), Ottawa, CA. p. 169-170. Gibson, R. W.; Mpembe, I.; Alicai, T.; Carey, E. E.; Mwanga, R. O. M.; Ngeve, J. M.; Boukamp, J. C. 1991. Seal, S. E.; Vetten, H. J. 1998a. Effects of sweetpotato virus disease Symptoms, aetiology and on the yield of sweetpotato serological analysis of sweetpotato genotypes in Cameroon. Exp. Agric. virus disease in Uganda. Plant 27:221-225. Pathol. 47:95-102. Rossel, H. W.; Thottappilly, G. 1987. Gibson, R. W.; Jeremiah, S. C.; Msabaha, Complex virus diseases of R. P. 1998b. Virus diseases of sweetpotato. In: Exploration, sweetpotato and cassava in the maintenance and utilization of Lake Zone of Tanzania. Report no. sweetpotato genetic resources. 2380, Natural Resources Institute Centro Internacional de la Papa (NRI), Chatham, GB. (CIP), Lima, Peru. p. 291-302.

Gibson, R. W.; Jeremiah, S. C.; Aritua, V.; Schaefers, G. A.; Terry, E. R. 1976. Insect Msabaha, R. P.; Mpembe, I.; transmission of sweetpotato Ndunguru, J. 2000. Sweetpotato diseases in Nigeria. Plant Pathol. virus disease in sub-Saharan 66:642-645. Africa: Evidence that neglect of seedlings in the traditional farming Sheffield, F. M. L. 1953. Virus diseases of system hinders the development of sweetpotato in parts of Africa. Emp. superior resistant landraces. J. Exp. Agric. 21:184-189. Phytopathology 148:441-447. Sheffield, F. M. L. 1957. Virus diseases of Hahn, S. K. 1979. Effects of sweetpotato sweetpotato in East Africa. virus disease (SPVD) on growth and 1. Identification of viruses and their yield of sweetpotato. Exp. Agric. insect vectors. Phytopathology 15:253-256. 47:582-587.

97 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 1.14 Conclusions and Recommendations

James Legg* and Braima James**

Introduction of cassava (Manihot esculenta Crantz) and sweetpotato (Ipomoea batatas [L.] The first phase of the Tropical Whitefly Lam.). Integrated Pest Management (TWF-IPM) Project (see Introduction to this volume) provided a unique opportunity Increased Biological to assess whitefly and whitefly-borne Understanding virus problems in a diverse range of cropping systems across the tropics and to obtain a broad range of Whitefly species and diagnostic information as a basis for abundance further concerted action. Sub-Project 4, Field data collection and the Whiteflies as virus vectors in cassava examination of whitefly nymph and sweetpotato in sub-Saharan Africa, specimens collected by project partners had the ambitious target of during surveys of cassava and establishing collaborative research sweetpotato in participating countries linkages among more than 15 partner provided the most comprehensive institutions and implementing information ever obtained for whiteflies diagnostic surveys in nine African on these crops in Africa. More than countries. Activities began in 2000 specimens were identified and September 1997 and were completed in four species were recorded. The two mid-1999. This chapter reviews the species that have been reported results obtained, considers the previously on cassava, Bemisia tabaci experiences gained in setting-up, (Gennadius) and B. afer (Priesner and implementing and completing the Hosny), were the only ones occurring diagnostic phase of research, and widely on cassava and sweetpotato and discusses the implications for were recorded from most locations. Two subsequent work on enhancing the species were identified that have not management of whitefly-borne diseases been reported previously from cassava. These were the greenhouse whitefly Trialeurodes vaporariorum Westwood, * International Institute of Tropical identified from two locations in Ghana Agriculture-Eastern and Southern Africa Regional Center (IITA-ESARC), Kampala, and Nigeria, and T. ricini (Misra) Uganda, and Natural Resources Institute, identified from a single site in the Lake University of Greenwich, Chatham Maritime, Victoria zone of Tanzania. Whilst the Kent, UK. ** IITA, Biological Control Center for Africa, occurrence of late instar nymphs of Cotonou, Benin. these two species does suggest

98 Conclusions and Recommendations colonization of cassava, the very low issues, particularly since the current frequency of occurrence means that surveys were conducted only at one they are of little significance. time in the year and therefore may not be fully representative. It is B. tabaci and B. afer both occurred important to highlight, however, that on cassava and sweetpotato, although B. afer has never been shown to B. afer was relatively less frequent on transmit viruses. The distinction sweetpotato and in Uganda was not between the two species therefore recorded from this crop. There was an becomes important if errors are not to important geographical zonation in the be made in describing relationships relative abundance of the two species between adult whitefly populations (Figures 1-3). Whilst B. tabaci and the spread of viruses causing predominated in countries in the cassava mosaic disease (CMD) or equatorial belt (Ghana, Benin, Nigeria, sweetpotato virus disease (SPVD). Cameroon, Uganda and Kenya), the two species occurred at similar Geographical patterns of variation frequencies in Tanzania, to the south, in abundance of cassava whiteflies and B. afer predominated in the most were partly confounded with the southerly countries, Malawi and different patterns of distribution of Madagascar. The reasons for this the two species (Figures 1-3). pattern of occurrence are not clear, Although survey teams were trained particularly since no single to differentiate the adult stages of the environmental factor appears to two species, it was not always correlate consistently with the species possible for them to do so during ratio. In Malawi, however, there was a counting. Some patterns in significant difference in the ratio distribution were apparent, however. between survey areas. In the two In West Africa, where B. tabaci was survey areas (1 and 2) along the shore the dominant species, abundance was of Lake Malawi (475-700 m altitude), generally greater in the more humid similar numbers of the two species rainforest and transition forest were recorded; whilst in survey area 3, ecozones. Conversely, populations the central plateau (1000-1500 m were smaller in the drier northern altitude), 97% of specimens identified savannah zones. This is a pattern were B. afer (Chapter 1.9, this volume). that has been recognized before, both in the Ecologically Sustainable The key difference between the Cassava Plant Protection (ESCaPP) lakeshore and plateau environments is surveys of 1993-94 (IITA, 1998) and temperature and it seems likely that through earlier studies of CMD the low temperatures that occur in the epidemiology in Ivory Coast (Fauquet mid-altitude environments of southern et al., 1987; 1988). In East Africa, two Africa are unfavourable for B. tabaci. centres of whitefly abundance were This possibility is supported by recognized: the northern shoreline of published information indicating that Lake Victoria, in Uganda, and the the development period of this species northern Tanzania/southern Kenya is significantly increased (Gerling et al., coast. As for West Africa, these are 1986) and fecundity significantly also the areas in the region where reduced (Azab et al., 1971) at humidity is greatest. It is notable, temperatures below 15 ºC. More however, that abundance of B. tabaci detailed ecological studies and in the “Lake Crescent” agro-ecological molecular characterization of the two zone in Uganda, comprising the species will be required to clarify these southern districts of Iganga, Mukono,

99 Whiteflies and Whitefly-borne Viruses in the Tropics 1-3 Not surveyed 40%-50% 50%-60% 60%-70% 70%-80% 3-9 9-27 27-81 20%-30% 30%-40% 0-1 Whitefly abundance CMD incidence Cameroon ratio for Sub-Project 4 countries in West Africa. Nigeria Bemisia tabaci/B. afer B. tabaci Benin B. afer Kilometers 0 200 400 600 Ghana 200 Figure 1.Figure Whitefly abundance, cassava mosaic disease (CMD) incidence and

100 Conclusions and Recommendations

Whitefly abundance 0-1

1-3

3-9 Uganda 9-27

Kenya 27-81

CMD incidence

Not surveyed

20%-30% 30%-40%

40%-50%

50%-60%

60%-70%

Tanzania

B. afer

B. tabaci

Malawi

100 0 100 200

Kilometers

Figure 2. Whitefly abundance, cassava mosaic disease (CMD) incidence and Bemisia tabaci/B. afer ratio for Sub-Project 4 countries in East Africa.

101 Whiteflies and Whitefly-borne Viruses in the Tropics

Whitefly abundance 0-1

1-3

3-9

9-27

27-81

CMD incidence

Not surveyed

20%-30% 30%-40%

40%-50%

50%-60% Madagascar 60%-70%

B. tabaci

B. afer

100 0 100 200 300 400

Kilometers

Figure 3. Whitefly abundance, cassava mosaic disease (CMD) incidence and Bemisia tabaci/B. afer ratio in Madagascar.

102 Conclusions and Recommendations

Mpigi, Masaka and Rakai, is recorded as participating countries to the John having increased significantly during the Innes Centre (JIC) for diagnostics and second half of the 1990s (Legg, 1995; characterization. Chapter 1.11 (this Otim-Nape et al., 1997; Legg and Ogwal, volume) reports the results of this 1998). This is now thought to have work. Until 1996, it was considered resulted from a synergistic interaction that two cassava mosaic with an unusually severe form of CMD, begomoviruses were occurring in which spread through Uganda during Africa, with largely disjunct this period (Colvin et al., 1999) and distributions (Swanson and Harrison, continues to spread in the East and 1994). African cassava mosaic virus Central African region (Legg, 1999). The (ACMV) was reported to occur in much interaction has occurred in all agro- of West, Central and southern Africa, ecologies in Uganda but it remains to be and East African cassava mosaic virus seen if the same will occur in the drier (EACMV) was said to be restricted to agro-ecologies to the south, in Tanzania. coastal East Africa, Madagascar, In Madagascar, there was no clear Malawi and Zimbabwe. More recent pattern of variation in abundance diagnostic surveys have shown that (Figure 3), although abundance in the EACMV is not restricted to East Africa southern central areas was lower than (Offei et al., 1999; Ogbe et al., 1999), elsewhere. that the distributions of EACMV and ACMV overlap (Ogbe et al., 1996; 1997) Variation in abundance of and that mixed virus infections occur sweetpotato whiteflies was much greater (Harrison et al., 1997). Results from than for cassava whiteflies (Table 1). the TWF-IPM Project (Chapter 1.11, Very high populations were recorded this volume) have extended our around the shores of Lake Victoria, understanding further by establishing whilst whiteflies were not recorded at all the first record of EACMV from Benin from many locations in Madagascar. It in West Africa; identifying a novel is likely that seasonal effects can partly begomovirus from cassava in explain this degree of variation, since Zimbabwe, which appears to be closely sweetpotato is grown for a much shorter related to South African cassava mosaic period (typically 4-6 months) than virus (Rey and Thompson, 1998); and cassava (typically 10-24 months). Thus, highlighting the occurrence of virus although cassava crops may be available mixtures in East Africa. Dual infections throughout the year, in drier areas, are a relatively uncommon occurrence sweetpotato cultivation outside the main (< 10%) and it is significant that all growing season is confined to very small these were recorded from the part of areas used for the maintenance of East Africa affected by the pandemic of planting material. Associated with this, unusually severe CMD (Otim-Nape et there is evidence showing that B. tabaci al., 1997; Legg 1999; Chapter 1.1, this on sweetpotato colonize a wide range of volume). crop plants and weed species, in contrast to B. tabaci on cassava, which Although virus diagnoses have appear to be largely restricted to been carried out on only a limited cassava (Burban et al., 1992; Legg et al., number of samples, a relationship is 1994). apparent between the viruses affecting cassava plants and the disease caused. Whitefly-transmitted viruses in In West Africa, diversity seems to be cassava limited and symptoms of disease are Stem samples of CMD-diseased cassava generally mild to moderate (Table 1). In were sent from all nine principal East and southern Africa, by contrast,

103 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. Summary of data on the incidence of cassava mosaic disease (CMD), sweetpotato virus disease (SPVD) and abundance of Bemisia tabaci, from Sub-Project 4 diagnostic surveys.

Country Survey area Fieldsa CMD Wfinfc CMD Cassava SPVD SP (%)b (miu) sev.d WFe (%)f WFg

Uganda North 20 (20) 77 48.0 3.0 1.2 5.0 1.40 East 20 (20) 68 46.0 3.1 1.8 12.0 5.40 South/Central 20 (20) 79 53.0 2.8 3.8 15.0 12.40 West 20 (20) 48 55.0 2.8 9.2 10.0 3.00

Kenya Coastal Province 17 (15) 56 36.0 2.3 2.9 0.5 0.50 Western Province 15 (18) 85 64.0 2.9 0.3 6.0 2.30 Nyanza Province 18 (17) 11 2.3 2.2 0.5 7.0 4.90

Madagascar Central plateau 35 (11) 31 7.8 2.6 7.1 3.0 0.01 Humid coast 22 (15) 71 11.0 3.5 4.3 2.0 0 Dry coast 12 (7) 46 11.0 3.2 5.9 18.0 0.10 Semi-humid coast 42 (6) 41 11.0 3.3 2.6 0 0

Tanzania Lake zone plateau 20 (20) 21 1.7 2.6 0.7 11.0 40.10 Semi-humid coast 20 (20) 12 7.1 2.7 1.2 13.0 2.80 Humid coast 20 (20) 39 21.0 3.0 5.8 6.0 3.10

Malawi Central lakeshore 20 62 55.0 2.8 1.5 - - Northern lakeshore 9 52 34.0 2.7 1.0 - - Central plateau 12 11 3.7 3.0 1.3 - -

Ghana Coastal savannah 10 76 44.0 2.2 11.8 - - Rainforest 25 69 23.0 2.2 1.5 - - Transition forest 25 74 54.0 2.5 1.0 - - Guinea savannah 20 68 44.0 2.4 0.3 - -

Benin Transition forest 28 26 3.4 2.1 4.2 - - Wet savannah 19 43 0.5 2.1 0.7 - - Dry savannah 13 49 3.4 2.1 3.4 - -

Nigeria Rainforest 25 83 3.9 2.4 3.1 - - Transition forest 25 32 2.7 2.1 2.6 - - Wet savannah 20 56 0 2.4 1.0 - - Dry savannah 10 45 2.4 2.2 0.3 - -

Cameroon South-west 26 54 21.0 2.2 4.1 - - North-west 16 55 21.0 2.4 2.4 - - Centre-south 28 73 5.6 2.3 3.1 - - a. Fields: Figures in parentheses indicate number of sweetpotato fields sampled. b. CMD %: Total incidence of CMD, calculated as the average percentage of plants showing symptoms of virus disease from 30 plant samples recorded in each dataset (i.e., one field). c. Wfinf (miu): Incidence of plants infected with CMD whitefly in the current season, transformed to allow for multiple infection units (Gregory, P. H. 1948. The multiple infection transformation. Ann. Appl. Biol. 35: 412-417). d. CMD sev.: Severity of CMD symptoms, based on an arbitrary 1-5 scale, where 1 indicates low and 5 indicates high severity of symptoms. e. Cassava WF: Mean whitefly abundance for cassava recorded as the average number of adult whiteflies counted on the five uppermost leaves of 30 plant samples recorded in each field. f. SPVD %: Total incidence of SPVD, calculated as the average percentage of plants showing symptoms of virus disease from 30 plant samples recorded in each field. g. SP WF: Mean whitefly abundance for sweetpotato recorded as the average number of adult whiteflies counted in ten 1-minute counts made at random in the sampled field.

104 Conclusions and Recommendations complexity appears greater, dual Whitefly-transmitted viruses in infections are more prevalent and sweetpotato severe disease is more widespread, Sweetpotato leaf samples with although in most instances (4/5) the symptoms of virus disease from dual infections were in areas affected Uganda, Kenya, Tanzania and Zambia by the East African CMD pandemic were analysed. Chapter 1.12 discusses (Legg, 1999). An association between in detail the results of virus diagnoses virus infection and epidemiology is from Uganda, Kenya and Tanzania, and similarly apparent (Table 1). In East Kaitisha and Gibson (1999) discuss Africa, current-season whitefly-borne those from Zambia. The East African infection is greatest in Uganda, where strain (S ) of Sweetpotato chlorotic the “Uganda variant” of EACMV EA stunt virus (SPCSV) occurred (EACMV-Ug) was reported (Chapter throughout the region and, in Uganda, 1.11, this volume) and where EACMV- Aritua et al. (Chapter 1.12, this Ug/ACMV mixtures were common, as volume) have discussed distributions reported earlier by Harrison et al. for the two distinct serotypes (SEA1 and (1997); whilst in Nigeria, where only S ) described by Alicai et al. (1999). ACMV was reported, current-season EA2 The other whitefly-borne virus whitefly-borne infection was very identified from project surveys was infrequent. Sweetpotato mild mottle virus (SPMMV), although this was restricted to areas The incidence of CMD varied around the shores of Lake Victoria considerably within sub-regions. In (Chapter 1.12, this volume). SPVD, West Africa, disease incidence ranged which results from co-infection of from 36% (Benin) to 72% (Ghana); sweetpotato plants with SPCSV and the whilst in East and southern Africa, it aphid-borne Sweetpotato feathery mottle ranged from 21% (Tanzania) to 68% virus (SPFMV), occurred throughout (Uganda). In all countries, the areas sampled. The disease was most proportion of diseased plants that had frequent in south-western Uganda, the arisen from planting diseased cuttings Lake Victoria zone of Tanzania, south- (cutting infection) was greater than that eastern coastal Tanzania and the semi- derived from current season whitefly- arid south-western part of Madagascar. borne infection. Many factors influence Diagnostic tests related SPVD to the CMD incidence, including the virus(es) co-occurrence of SPCSV and SPFMV infecting the plant, cultivar response to for each of these regions (Chapter 1.12, the virus(es), vector abundance and this volume). No samples from transmission efficiency, the physical Madagascar were tested. environment, and cropping practices (including vector or disease Whitefly natural enemies management measures). Most of these factors remain poorly defined for many The agreed survey protocol for the of the participating countries, although TWF-IPM Project provided guidelines the surveys have provided quantitative for the collection of parasitoids, data relating to some aspects. In this predators and entomopathogens during respect it would be useful to compare the course of diagnostic surveys. the epidemiology of CMD between Limitations in the methods proposed, regions, using common cultivars for and the short period of time available reference, as proposed by Legg et al. for natural enemy collection at each (1997). sampling site, meant that the collection of natural enemies was less comprehensive than had been

105 Whiteflies and Whitefly-borne Viruses in the Tropics anticipated. More intensive and Burban, 1994). Whilst this component targeted surveys would be required for of the TWF-IPM Project’s diagnostic the adequate characterization of phase had deficiencies, it is predators and entomopathogens in nevertheless considered that the particular. information obtained has provided an important baseline from which to Survey teams in most countries develop future studies. Key issues that collected parasitized whitefly mummies will need to be more fully addressed and reared adult parasitoids from include the variation in parasitoid them. The parasitoids were diversity between regions, of which subsequently mounted and identified there was evidence in this study, and by Dr. Mohammed Ali Bob of the the role (if any) of the different International Center of Insect parasitoid species in the population Physiology and Ecology (ICIPE). dynamics of Bemisia spp. and the Dr. A. Polaszek of the British Museum epidemiology of CMD. (Natural History) verified representative specimens. Five species of parasitoid were recorded: Encarsia sophia (Girault and Dodd), Encarsia lutea (Masi), Increased Socio-Economic Encarsia mineoi Viggiani, Encarsia sp. Understanding (luteola group) and Eretmocerus sp. (Table 2). Only two species, E. sophia Farmers’ assessment of and Eretmocerus sp., were widely whitefly-related problems distributed, parasitizing whiteflies on both cassava and sweetpotato. There was a clear relationship between E. sophia was an order of magnitude farmers’ assessments of CMD and more common that Eretmocerus sp., SPVD and the prevalence and severity although it is considered that easy of the diseases as described from the recognition of the former, resulting field data collection. In Uganda, all from the black coloration of the farmers both recognized CMD and mummy from which it emerges, may considered it a problem for their have partially biased the sampling. cassava production. In the countries of Whilst over 30 species of parasitoid West Africa, on the other hand, have been described as parasitizing although typically about 70% of B. tabaci worldwide (Gerling, 1986), farmers were able to recognize CMD, only limited information is available less than 50% considered it a problem. about Bemisia parasitoids in Africa, A smaller proportion of sweetpotato and virtually no information relating to farmers recognized SPVD but, where it cassava or sweetpotato (Fishpool and was recognized, most farmers

Table 2. Aphelinid parasitoids identified from diagnostic surveys of Sub-Project 4.

Country Encarsia Eretmocerus Encarsia sp. Encarsia Encarsia sophia sp. (luteola group) lutea mineoi

Uganda 62 46 3 - - Kenya 13 2 - - - Tanzania 62 - 1 - - Ghana 4 - - - - Benin 16 2 - - - Nigeria 557 21 4 7 5 Cameroon 324 38 - - -

106 Conclusions and Recommendations considered it a problem. Much smaller reported doing anything at all to proportions of farmers were able to control CMD. This result was recognize whiteflies and many particularly anomalous for Ghana, expressed surprise when informed of since many farmers reported relatively their role as disease vectors. There high losses. Management practices were significant differences between were most widely reported in Uganda countries in this respect, however, with and Cameroon, where most farmers the extremes being Madagascar, where indicated the use of at least one no farmers reported recognizing practice. The most cited methods of whiteflies, and Benin, where 70% disease management were the recognized them. phytosanitary techniques of roguing, and selection of disease-free stems. Farmers had names for whiteflies This has to be set against the and whitefly-transmitted diseases in all predominance of the use of diseased countries, but in many cases these cuttings as planting material in almost were non-specific, translating simply as all areas surveyed. Selection of disease- “insect” or “disease”. More specific free planting material similarly was names were commonest for CMD, cited widely as a management tactic which was often described using but the health status of stems was anthropomorphic disease equivalents normally less important as a criterion such as “leprosy”. Names for SPVD for planting material selection than were often exactly the same as those were agronomic characters. used for CMD. Where such names are in common usage, potential exists for Although it appears that farmers their use, perhaps with minor over-estimated their use of roguing and modification, in training and extension selection of healthy cuttings, they programs in the future. Estimates of under-reported use of disease-resistant loss to the respective diseases were varieties. Use of cassava varieties most frequently in the range of 25%- recommended for resistance to CMD 75% for cassava and 25% or less for was most widespread in Cameroon, sweetpotato, although most survey where 48% of farmers reported growing teams considered that these were likely variety “Agric” in the knowledge that it to be overestimates, particularly in was resistant to CMD. In Uganda, more West African countries, where disease than 15% of sweetpotato farmers were symptoms (in this case only CMD) were found to be growing SPVD-resistant typically mild. Participatory yield loss varieties but none of these realized that studies would be helpful in many of the this was a method of managing SPVD. surveyed countries to evaluate losses scientifically and reconcile such Very few farmers reported using estimates with farmer perceptions. The pesticides to control whiteflies or shortcoming of this and other whitefly-borne viruses. Occasionally, responses emphasized the need to link sweetpotato farmers reported pesticide producer interviews with the collection use but this was more likely to be for of more objective field data. the purpose of controlling other insect pests such as the sweetpotato butterfly Managing whiteflies and Acraea acerata Hewitson. Chapter 1.13 whitefly-transmitted viruses (this volume) gives a more detailed assessment of the approaches of There were widely divergent responses sweetpotato farmers to SPVD control. on the use of vector and disease Few farmers reported having received management measures. In Ghana and any technical assistance with respect Madagascar, less than 5% of farmers

107 Whiteflies and Whitefly-borne Viruses in the Tropics to the control of either CMD or SPVD disease-resistant varieties, selecting and, as a result, most considered that disease-free planting material and/ the management methods being used or roguing of diseased plants); and were their own. The only exception to this was Cameroon, where 39% of (3) Facilitate the adoption of these farmers reported that they had received methods by using participatory technical support relating to whiteflies training and all media channels and whitefly-borne diseases. available to improve the access of farmers to current knowledge.

Recommendations Sub-Project 4 of the TWF-IPM Project has been successful in The overall picture that emerges from achieving part of the first objective. this assessment of farmers’ awareness Following the completion of the of whiteflies and whitefly-borne diagnostic phase, there is a clearer diseases in cassava and sweetpotato understanding of the distribution of and their approach to disease whitefly-borne viruses, what their management is one of weak impact is in terms of disease incidence understanding of the nature of the and epidemiology and how farmers are problem and a correspondingly responding to these diseases. The haphazard implementation of project also has been uniquely management tactics. Much of the successful in obtaining comparable responsibility for this lies with the information of this type from nine of the research (both national and major cassava-producing countries of international) and extension systems. Africa and four of the major Researchers have failed to provide a sweetpotato-producing ones; these well-defined and well-tested strategy for countries together account for more managing CMD and SPVD. Moreover, than 50% of African production of each weakness in the extension systems of crop. In achieving this, the project has most of the participating countries has established a network of researchers meant that even where management working on whiteflies and whitefly- recommendations have been defined, borne disease of root crops in sub- these are not reaching the majority of Saharan Africa. In order to complete farmers (as is particularly the case for the first objective and to begin to the control of whitefly-borne diseases of address the second, the following cassava and sweetpotato in Uganda). In issues, a number of which have been order to address this situation, key highlighted in the country chapters, objectives for future work on whiteflies need to be addressed. and whitefly-borne diseases, for both cassava and sweetpotato, should be to (1) How severe are yield losses in the encourage researchers, extension CMD-diseased local landraces that agents and farmers to work together to: still dominate cassava production in Africa? Knowledge of this is (1) Define exactly where disease particularly critical, for example, in management is needed and where Ghana, where farmers report high not; losses yet appear to make little (2) Develop effective, clearly defined attempt to control them. A realistic and readily implemented vector estimate of current yields and yield and disease management methods losses would help direct research (which based on current knowledge and extension efforts more are most likely to include choosing effectively.

108 Conclusions and Recommendations

(2) What are the conditions under addition to technical issues; the first which tactics for improving plant phase, described in this volume, has health, especially selection of provided some important lessons in disease-free cuttings and roguing this regard. of diseased plants, are most likely to be effective? (1) The coordination structure developed during the first phase (3) Where is host plant resistance has provided an effective means by required as a foundation for which a global-scale project can be sustainable management of executed. Effective interaction, cassava and/or sweetpotato (as facilitated by electronic compared with a strategy focusing communication, is vital for the on improving the husbandry of efficient working of this structure. land races) and are particular virus/vector/environmental (2) At the outset of the next phase of characteristics common to these the project, all partners at sub- areas? project level should participate in an introductory meeting, during (4) Where conditions demand the use which opportunities are provided of host plant resistance, as has for critical review of the proposed been the case in Uganda during the work plans and budgets, and for pandemic of severe CMD, what are training in additional skills the constraints to the adoption of required for project newly developed CMD-resistant implementation. This was not done varieties? during the first phase and, as a result, it was difficult to (5) Are there alternative, novel consistently train partners in approaches to crop health national research systems and to management that can be used to provide all partners with an supplement existing methods? appreciation of the continental and Although both biological control global contexts of the project. and host plant resistance are regarded conventionally as being (3) The sub-project should budget for more appropriate for use against annual review meetings for all direct pests, rather than virus principal partners. If resources are vectors, is there scope for their limited, the number of partners deployment? should be restricted to allow full participation of each. (6) Where are successes currently being realized in CMD and/or (4) If a large number of national SPVD management and how can research system partners are to be experience from these situations be included, as was the case in the incorporated into research and first phase of Sub-Project 4, the extension efforts elsewhere? task of co-ordination and provision of research support, executed by The second phase of the TWF-IPM the co-ordinating international Project proposes to address key centre, should be divided between research and implementation issues major regions covered. In the first such as those indicated above. In the phase of Sub-Project 4, the formulation of this vitally important International Institute of Tropical phase of the project, attention needs to Agriculture (IITA) provided co- be paid to collaborative mechanisms in ordination and support through

109 Whiteflies and Whitefly-borne Viruses in the Tropics

offices in both West and East References Africa. This division of responsibility was considered to be Alicai, T.; Fenby, N. S.; Gibson, R. W.; a key factor in the successful Adipala, E.; Vetten, H. J.; Foster, G. D.; Seal, S. E. 1999. Occurrence of implementation of the sub-project two serotypes of Sweetpotato chlorotic although such a division demands stunt virus in East Africa and their an effective exchange of ideas and associated differences in coat information between the regions. protein and HSP70 homologue gene sequences. Plant Pathol. Whiteflies continue to pose a 48:718-726. tremendous challenge to farmers, researchers and development workers Azab, A. K.; Megahed, M. M.; El-Mirsawi, in Africa, as well as in the wider global D. H. 1971. On the biology of Bemisia tabaci (Genn.). Bull. Soc. context. However, significant successes Entomol. Egypte 55:305-315. are being realized in meeting this challenge both in terms of Burban, C.; Fishpool, L. D. C.; Fauquet, characterizing the problems and in C.; Fargette, D.; Thouvenel, J. -C. developing strategies to address them 1992. Host associated biotypes in a sustainable manner. In the African within West African populations of the whitefly Bemisia tabaci (Genn.) context, at stake are the livelihoods of (Hom., Aleyrodidae). J. Appl. many millions of producers, who Entomol. 113:416-423. depend on cassava and sweetpotato as sources of cash income, and Colvin, J.; Otim-Nape, G. W.; Holt, J.; consumers, both rural and urban, who Omongo, C.; Seal, S.; Stevenson, P.; depend on these crops to provide Gibson, R.; Cooter, R. J.; Thresh, J. affordable staple foods. It is to be M. 1999. Factors driving the current epidemic of severe cassava mosaic hoped, therefore, that initiatives to disease in East Africa. In: Procs. strengthen management of whitefly- VIIth International Plant Virus transmitted virus diseases of these two Epidemiology Symposium, 11-16 major food crops, including the second April 1999, Almeria, ES. p. 76-77. phase of the TWF-IPM Project, will be able to attract a level of support Fauquet, C.; Fargette, D.; Thouvenel, commensurate with the magnitude of J. -C. 1987. Epidemiology of African cassava mosaic on a regional scale the problem. in the Ivory Coast. In: Procs. International Seminar on African cassava mosaic disease and its Acknowledgements control, 4-8 May, Yamoussoukro, CI. Centre Technique de Coopération A grant from the Danish International Agricole et Rurale, Wageningen, NL, Development Assistance funded the and Institut Français de Recherche diagnostic phase of the TWF-IPM Scientifique pour le Développement en Coopération, Paris, FR. Project, Sub-Project 4. Each of the p. 131-135. project partners provided complementary support. The authors Fauquet, C.; Fargette, D.; Thouvenel, would like to extend particular thanks J. -C. 1988. Some aspects of the to Dr. Mohammed Ali Bob of ICIPE, epidemiology of African cassava who identified all the whitefly and mosaic virus in Ivory Coast. Trop. parasitoid samples for Sub-Project 4. Pest Manag. 34:92-96. Fishpool, L. D. C.; Burban, C. 1994. Bemisia tabaci: The whitefly vector of African cassava mosaic geminivirus. Trop. Sci. 34:55-72.

110 Conclusions and Recommendations

Gerling, D. 1986. Natural enemies of Legg, J. P.; Gibson, R. W.; Otim-Nape, G. Bemisia tabaci, biological W. 1994. Genetic polymorphism characteristics and potential as amongst Ugandan populations of biological control agents: A review. Bemisia tabaci (Gennadius) Agric. Ecosyst. Environ. 17:999-110. (Homoptera: Aleyrodidae), vector of African cassava mosaic geminivirus. Gerling, D.; Horowitz, A. R.; Baumgärtner, Trop. Sci. 34:73-81. J. 1986. Autecology of Bemisia tabaci. Agric. Ecosyst. Environ. 17:5-20. Legg, J. P.; James, B.; Cudjoe, A.; Saizonou, S.; Gbaguidi, B.; Ogbe, Gregory, P. H. 1948. The multiple infection F.; Ntonifor, N.; Ogwal, S.; Thresh, transformation. Ann. Appl. Biol. J.; Hughes, J. 1997. A regional 35:412-417. collaborative approach to the study of ACMD epidemiology in sub- Harrison, B. D.; Zhou, X.; Otim-Nape, G. Saharan Africa. Afr. Crop Sci. J. W.; Liu, Y.; Robinson, D. J. 1997. 3:1021-1033. Role of a novel type of double infection in the geminivirus-induced Ogbe, F. O.; Songa, W.; Kamau, J. W. epidemic of severe cassava mosaic in 1996. Survey of the incidence of Uganda. Ann. Appl. Biol. 131: African cassava mosaic and East 437-448. African cassava mosaic viruses in Kenya and Uganda using a IITA (International Institute of Tropical monoclonal antibody based Agriculture). 1998. Ecologically diagnostic test. Roots 3:10-13. Sustainable Cassava Plant Protection (ESCaPP) in South America and Ogbe, F. O.; Legg, J.; Raya, M. D.; Africa. Africa: Project findings and Muimba-Kankolongo, A.; Theu, M. recommendations. Ibadan, NG. P.; Kaitisha, G.; Phiri, N. A.; Chalwe, A. 1997. Diagnostic survey of Kaitisha, G. C.; Gibson, R. W. 1999. cassava mosaic viruses in Tanzania, Identification of sweetpotato feathery Malawi and Zambia. Roots 4:12-15. mottle and chlorotic stunt viruses in sweetpotato in Zambia. In: Akoroda, Ogbe, F. O.; Atiri, G. I.; Robinson, D.; M. O.; Teri, J. M. (eds.). Procs. Winter, S.; Dixon, A.; Quin, F. M.; Scientific Workshop of the Southern Thottappilly, G. 1999. First report of African Root Crops Research Network East African cassava mosaic (SARRNET), 17-19 August 1998, begomovirus in Nigeria. Plant Dis. Lusaka, ZM. p. 293-299. 83:398.

Legg, J. P. 1995. The ecology of Bemisia Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; tabaci (Gennadius) (Homoptera: Baguma, Y.; Ogwal, S.; Semakula, Aleyrodidae), vector of African G. N.; Acola, G.; Byabakama, B.; cassava mosaic geminivirus in Martin, A. 1997. Cassava mosaic Uganda. Ph.D. thesis, University of virus disease in Uganda: The Reading, GB. current pandemic and approaches to control. Natural Resources Legg, J. P. 1999. Emergence, spread and Institute (NRI), Chatham, GB. 65 p. strategies for controlling the pandemic of cassava mosaic virus Rey, M. E. C.; Thompson, G. 1998. disease in east and central Africa. Cassava mosaic virus disease in Crop Prot. 18:627-637. South Africa. Roots 5:3-5.

Legg, J. P.; Ogwal, S. 1998. Changes in the Swanson, M. M.; Harrison, B. D. 1994. incidence of African cassava mosaic Properties, relationships and geminivirus and the abundance of its distribution of cassava mosaic whitefly vector along south-north geminiviruses. Trop. Sci. 34:15-25. transects in Uganda. J. Appl. Entomol. 122:169-178.

111 BLANCA 112 SECTION TWO

Whiteflies as Pests and Vectors of Viruses in Vegetable and Legume Mixed Cropping Systems in Eastern and Southern Africa BLANCA 114 Introduction

CHAPTER 2.1 Introduction

Mohamed Ali Bob, Rebbeca Raini and Bernhard Löhr*

In some parts of Africa such as in the Selection of Target Crops Sudan and Tanzania the countryside and Target Countries resembles a graveyard, not through the ravages of civil war or natural disaster Tomato is one of the most widely but because of one pest, the whitefly cultivated vegetable crops in the region, Bemisia tabaci (Gennadius), which is predominantly grown by small-scale sucking the life out of Africa’s crops. producers for fresh consumption (GTZ, Bemisia, which was originally only a 1995). National scientists have ranked problem in industrial crops such as B. tabaci as the most important pest/ tobacco (Nicotiana tabacum L.) and vector in tomato in Tanzania and cotton (Gossypium hirsutum L.) is now Kenya, and the second most important much more widespread and despite all in Malawi after red spider mite (Varela efforts to control this pest, it has and Pekke, 1995). In the Sudan, stayed one step ahead of man’s actions B. tabaci is considered the crop pest of against it. Bemisia has done this not highest economic importance only by developing resistance to many (Eveleens, 1983; Ahmed et al., 1987). of the chemicals commonly used but also by diversifying its tastes and its habitat to a much wider range of staple Knowledge Available on crops such as cassava (Manihot Whitefly-Related Problems esculenta Crantz) and common bean (Phaseolus vulgaris L.) to high-value Despite these opinions, no quantitative horticultural crops such as tomato documentation was available on the (Lycopersicon esculentum Mill.) and economic importance of whiteflies as melon (Cucumis melo L.). On-farm pests and/or vectors in vegetable surveys in the region have shown that cropping systems in the countries of the number of producers who have this sub-project prior to its abandoned their crops because of commencement. There was no clear tomato leaf curl diseases is on a sharp map of the areas affected in these increase. In the Sudan especially, countries and related information on whiteflies have forced entire areas out insecticide use and the cost of control of production, depriving producers of in the affected areas. Knowledge on food and cash income. whitefly-transmitted viruses in vegetables, the host range and natural enemies was also lacking in most partner countries before this project * International Center of Insect Physiology and began. Information on the status of Ecology (ICIPE), Nairobi, Kenya. whitefly resistance to insecticides was

115 Whiteflies and Whitefly-borne Viruses in the Tropics available only from the Sudan (see A progress meeting held at the Chapter 2.2). ICIPE followed this up in June 1998. Progress of the two African sub- Although the history of whitefly projects was presented, experiences research in the Sudan goes back to the shared and priority research areas for 1930s (Kirkpatrick, 1931), a Phase 2 were considered. The substantial collection of grey literature participating co-ordinators and on whitefly and whitefly-transmitted national teams made 10 presentations viruses from local stations/project on whitefly research. reports in the Sudan revealed that the research has focused almost entirely Finally, a research agenda for on whitefly-related problems in cotton Phase 2 was outlined at a joint production. Whitefly-related problems planning workshop held at the ICIPE in in vegetable production appear more September 1999. Research priorities recent, for example, Tomato yellow leaf were thoroughly discussed and priority curl virus (TYLCV) was first identified in setting was carried out with full 1965 (Yassin and Nour, 1965). participation of all national scientists of Nevertheless, research on whitefly- Sub-Project 4, co-ordinators and related problems in vegetable crops has collaborators from advanced research been limited. institutions.

In Tanzania, TYLCV was first identified in 1990 (Czosnek et al., Continent-Wide Appeal 1990). Later, Chiang et al. (1997) identified a different virus, named Additionally, the formation of an ToLCV-Tz, which has complicated the African Whitefly and Geminivirus scenario in Tanzania. No further Network was spearheaded by the sub- documentation on whitefly-related project at the 12th Meeting of the problems in vegetable crops is available African Association of Insect Scientists from this country. In Kenya and held in June 1997 in Stellenbosch, Malawi, documentation on whitefly- South Africa. Thirty-two researchers related problems is only available on from 12 countries in Africa participated citrus (Citrus spp. L.) and cassava. and expressed interest in participation in a regional information exchange network led by ICIPE. Planning and Co-ordination References In order to form the basis for the diagnostic activities for the first phase, Ahmed, A. H. M.; Elhang, E A., Bashir, H. H. 1987. Insecticide resistance in a planning and methodology workshop the cotton whitefly (Bemisia tabaci was held, with the participation of Genn.) in the Sudan Gezira. Trop. scientists from the national teams, at Pest Manag. 33:67. the International Center of Insect Physiology and Ecology (ICIPE) in April Chiang, B. T.; Nakhla, M. K.; Maxwell, P.; 1997. A detailed work plan for the first Schoenfelder, M.; Green, S. K. phase of the project was developed, 1997. A new geminivirus associated with a leaf curl disease of tomato in methodologies were discussed and Tanzania. Plant Dis. 81:111. views and experiences were shared among the participants.

116 Introduction

Czosnek, H.; Navot, N.; Laterrot, H. 1990. Kirkpatrick, T. W. 1931. Further studies Geographical distribution of Tomato on leaf-curl of cotton in the Sudan. yellow leaf curl virus. A first survey Bull. Entomol. Res. 22:323-363. using a specific DNA probe. Phytopathol. Mediterr. 29:1-6. Varela A. M.; Pekke, A. 1995. Proceedings of the Tomato Planning Workshop Eveleens, K. G. 1983. Chemical insect for Eastern and Southern Africa control in the Sudan Gezira: Region, 16-20 October 1995, Analysis of a crisis crop. Crop Prot. Harare, ZW. Deutsche Gesellschaft 2:273-287. für Technische Zusammenarbeit (GTZ)- Integrated Pest Management GTZ (Deutsche Gesellschaft für (IPM) Horticulture Project, Nairobi, Technische Zusammenarbeit). KE. 32 p. 1995. Major pests of tomatoes in eastern and southern Africa: Yassin, A. M.; Nour, M. A. 1965. Tomato Compilation of past research work leaf curl disease, its effects on yield and identification of IPM and varietal susceptibility. Sudan opportunities. GTZ-Integrated Pest Agric. J. 1:3-7. Management (IPM) Horticulture Project Report, Nairobi, KE. 32 p.

117 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 2.2 Sudan

Gassim Dafalla* and Musa Ahmed**

Introduction Wadi Halpha

Geographical context Argo Diagnostic surveys were conducted during 1997-99 following the Al Khurtum methodology agreed among project Berbetti partners, and covering Gezira, Butana, Dindir, Hasaheisa, Kamlin, Managil, Umulgra, White Nile, El Jabalain, Um Sudan Rawaba, Er Rahad, Dongola, Merewe, Gedaref and Nahr el Rahad in central, eastern, northern and western Sudan Pibor Post (Figure 1). Most of the surveyed sites Towns are low lying (< 500 m above sea level). Surveyed The climate is hot and dry for most of areas the year, with temperatures in summer Figure 1. Areas surveyed for whitefly incidence reaching more than 40 °C and relative in the Sudan. humidity less than 10%, and irrigation is commonly used. The driest of the survey areas are Dongola and Merewe economically most important crop pest in the Northern State. Gedaref region in the Sudan (Eveleens, 1983; Ahmed and some areas in Kordufan Province et al., 1987). In addition to the direct are at a somewhat higher altitude damage this whitefly causes through (> 500 m above sea level) and receive feeding, it disseminates a number of enough rainfall for rain-fed vegetable viruses that affect fibre and vegetable production. crops grown in the country.

The emergence of Bemisia tabaci Research on B. tabaci and whitefly- as a pest and virus vector transmitted viruses (WTVs) in Sudan The whitefly species Bemisia tabaci started in the late1920s, when (Gennadius) is considered to be the recurrent outbreaks of Cotton leaf curl virus (CLCuV), transmitted by B. tabaci, threatened to end cultivation of cotton * Plant Pathology Centre, Faculty of (Gossypium hirsutum L.) in the Gezira Agricultural Science, University of Gezira, (Kirkpatrick, 1931). Since then, a Wad-Medani, Sudan. ** Agricultural Research Corporation, considerable research effort has been Wad-Medani, Sudan. focused on B. tabaci and WTVs in

118 Sudan cotton, leading to the introduction of strategies, the problem has become CLCuV-resistant varieties in the Gezira more complicated and has defied in the late 1960s. However, research on vegetable producers’ efforts to whitefly-related problems in vegetable overcome it and secure a return for crops has been limited. Vegetables are their efforts. The need to design viable severely affected by WTVs and the integrated control strategies, based on incidence of WTVs in a given season a strong scientific foundation and constitutes the major factor knowledge of the field situation, has determining the success or failure of emerged as a major challenge for tomato (Lycopersicon esculentum Mill.), research. cucurbits, okra (Abelmoschus esculentus [L.] Moench), pepper Little information was available on (Capsicum annuum L.) and common the whitefly-related problems of bean (Phaseolus vulgaris L.) in that vegetable production in Sudan prior to season. It is estimated that, on the commencement of the Tropical average, 75% of the fruit yield of Whitefly Integrated Pest Management tomato, watermelon (Citrullus lanatus (TWF-IPM) Project. Many findings from [Thunb.] Matsum. & Nakai), musk the survey conducted in the first, melon (Cucumis melo L.) and okra is diagnostic phase of the project are new lost each year to WTVs. and add substantially to our knowledge base on biological and socio-economic Yassin and Nour (1965) first aspects of the whitefly and WTV reported Tomato leaf curl virus, problems of vegetable cropping systems transmitted by B. tabaci. Later in Sudan. investigations showed that the virus from Sudan is similar to Tomato yellow leaf curl virus (TYLCV) from the East Increased Biological Mediterranean region. In recent years, Understanding severe yellowing symptoms were observed in watermelon and melon grown in many parts of the country, Characterization of similar to those described for the begomoviruses and whitefly B. tabaci-transmitted Watermelon biotypes chlorotic stunt virus (WmCSV) in Yemen. Samples of whitefly adults and nymphs CLCuV not only affects cotton but also were collected mostly from tomato, with has been reported by many workers to a few from lablab (Lablab purpureus [L] cause heavy damage in okra in most Sweet) and Nalta jute (Corchorus parts of the country. Both cotton and olitorius L.), on 80 farms. One hundred okra belong to the mallow family. and fifty specimens were processed and According to some workers (cited by mounted on 76 slides and sent to the Dafalla and Sidig, 1997) the whitefly- International Center of Insect transmitted African cassava mosaic Physiology and Ecology (ICIPE) for disease is still the most destructive in identification. All the mounted nymph cassava in Western Equatoria Province specimens were identified as B. tabaci. in southern Sudan. In an experimental field at the At producer level, research efforts University of Gezira, severe silver leaf have not been reflected sufficiently in symptoms were noticed on squash the successful management of B. tabaci (Cucurbita pepo L.), heavily infested and WTVs, especially in vegetables. In with whiteflies. These symptoms have the absence of such management been associated elsewhere with the

119 Whiteflies and Whitefly-borne Viruses in the Tropics

B. tabaci B-biotype (Shapiro, 1995) but whose identity is yet to be established, molecular characterization of these were also noticed on pepper. whitefly specimens is still pending. In general, some variability has been Some weeds, Acalypha indica L., noticed in the patterns of behaviour Datura stramonium L. and Solanum and efficiency in virus transmission, coagulans Forskål, were identified as which also may be circumstantial TYLCV reservoirs. Many other non- evidence of the existence of different cultivated host plants of B. tabaci may Bemisia biotypes. act as WTV reservoirs. Continuous cultivation of tomato and other Host plants on which whiteflies solanaceous vegetables, coupled with were found to reproduce, identified in the abundance of alternative hosts for the course of surveys in Sudan, vector and viruses, is very likely to belonged to the following families: contribute to the build up of disease Amaranthaceae, Asclepiadaceae, problems. Asteraceae, Brassicaceae, Commelinaceae, Convolvulaceae, Cucurbitaceae, Euphorbiaceae, Disease incidence and symptom Lamiaceae, Leguminosae, Malvaceae, severity Solanaceae, Tiliaceae and Verbenaceae The Gezira Irrigation Scheme is the area (Table 1). with the most severe and persistent WTV disease problems in the country. Many of the WTVs in vegetables in Incidences of 80-100% were frequently Sudan have been well characterized but recorded in Gezira, Hasaheisa, Managil, more information on the inter-seasonal Umelgura, Dindir and Butana. In half of variation of virus incidence and on the the tomato fields in the Gezira Province, distribution and economic importance 100% TYLCV incidence was recorded. of various WTVs in the different regions B. tabaci was found to be more is needed. The field surveys revealed abundant in the southern part of Gezira that TYLCV, WmCSV, Okra leaf curl that receives less rainfall than the virus and Pepper leaf curl virus are the northern region. High whitefly most widespread and economically populations and TYLCV incidence were important WTVs in Sudan. Tomato, observed in Central Sudan, significantly watermelon, melon, okra, pepper and lower populations and TYLCV common bean were among the most incidences (1%-9%) were observed in affected crops. Nahr el Rahad (Eastern State), Dongola and Merewe (Northern State), and in Other viruses detected include Umm Rawaba and Er Rahad (Western Tomato vein thickening virus, which State). No TYLCV symptoms were causes another important disease of observed in Gedaref Province (Babiker, tomato in Sudan and Bean mild mosaic Khor-Garab, El Ramla, Basunda, Doka) virus (BMMV). Cucumber vein yellowing and the Abu Habil Basin. Vegetable virus (CVYV) and a virus inducing gardens along the Blue Nile River were potyvirus-like yellowing have been once the sole suppliers of tomato observed in cucurbits but the latter is outside the period suitable for rain-fed yet to be identified. In legumes, a cultivation. This is no longer the case geminivirus and a Closterovirus that because of severe TYLCV epidemics in affect cowpea (Vigna unguiculata [L.] recent years. Walp.) and common bean have been observed and are yet to be identified. The overall impression provided by Symptoms of another putative WTV, the surveys is that B. tabaci does not

120 Sudan

Table 1. Reproductive host plants of Bemisia tabaci (Gennadius) in the Sudan.

Family Species Common name

Crops Brassicaceae Raphanus sativus L. Radish Cucurbitaceae Citrullus lanatus (Thunb.) Matsum & Nakai Watermelon Cucumis melo L. Melon Cucumis sativus L. Cucumber Cucurbita pepo L. Squash Euphorbiaceae Manihot esculenta Crantz Cassava Leguminosae Cajanus cajan (L.) Millsp. Pigeon pea Cassia alexandrina Mill. Senna Lablab purpureus (L.) Sweet Lablab Lens culinaris Medik. subsp. culinaris Lentil Phaseolus vulgaris L. Common bean Vicia faba L. Broad bean Lamiaceae Ocimum basilicum L. Basil Malvaceae Abelmoschus esculentus (L.) Moench Okra Gossypium hirsutum L. Cotton Hibiscus cannabinus L. Kenaf Solanaceae Capsicum annuum L. Pepper Lycopersicon esculentum Mill. Tomato Tiliaceae Corchorus olitorius L. Nalta jute

Non-cultivated hosts Amaranthaceae Achyranthes aspera L. Amaranthus tricolor L. Asclepiadaceae Leptadenia sp. Asteraceae Ageratum conyzoides L. Bidens pilosa L. Sonchus sp. Commelinaceae Commelina benghalensis L. Convolvulaceae Ipomoea cordofana Choisy Euphorbiaceae Acalypha indica L. Euphorbia aegyptiaca Boiss. Euphorbia heterophylla L. Leguminosae Rhynchosia hirta (Andrews) Meikle & Verdc. Malvaceae Abutilon pannosum (G. Foster) Schltdl. Solanaceae Datura stramonium L. Solanum incancum L. Solanum nigrum L. Verbenaceae Lantana camara L.

cause major problems outside of areas grown, whereas WmCSV is present in where cotton is cultivated. For almost all melon production areas. example, in the Abu Habil Basin, where no cotton is grown, tomato is typically produced without whitefly infestation. Natural enemy species Similarly, OLCV is present in many Seventeen collections of whitefly parts of the country where cotton is parasitoids and predators were made

121 Whiteflies and Whitefly-borne Viruses in the Tropics and preliminary identification made. carota L. subsp. sativus [Hoffm.] The most common parasitoids of Arcang. var. sativus Hoffm.) and leafy B. tabaci were found to be Encarsia vegetables. Most producers (82%) lutea (Masi) and Eretmocerus mundus practised crop rotation. Mercet. The following natural enemies that were found in the vegetable-based Whiteflies and/or WTVs were found systems in the country also could be to be the greatest cause of concern to involved in regulating B. tabaci most tomato producers in the survey. populations: the coccinellids Coccinella Other destructive pests reported were undecimpunctata L., Hippodamia Helicoverpa armigera (Hübner), variegata (Goeze), Cheilomenes Liriomyza sativae Blanchard, sulphurea (Olivier) and Scymnus sp.; Scrobipalpa sp. and Keiferia the lacewings Chrysoperla pudica lycopersicella (Walshingham). (Navás) and Chrysoperla spp.; pirate bugs, Orius sp.; other predatory true Most tomato producers in Sudan bugs Campylomma sp. and spiders. (75% overall, including all those in Central Sudan) were able to recognize whiteflies and even more (89%) Increased Socio-economic recognized TYLCV. However, 65% of Understanding them did not know that TYLCV and B. tabaci were inter-related. The majority (84%) believed that whiteflies Farmers’ assessment of and/or TYLCV were serious production whitefly-related problems constraints in their farms. Whereas More than 95% of vegetable production 65% of the producers reported that in the surveyed areas is carried out on TYLCV was a problem, only 35% small-scale holdings of 2 to 5 ha. reported that both the whitefly and the Among tomato producers in Sudan, disease were problematic, and none 40% farmed their own land, 38% used reported a problem with whitefly alone. rented land, while 22% were squatters. In contrast, most producers in Gedaref, Virtually all farmers interviewed (99%) Northern and North Kordufan were men. Almost all the producers Provinces, where TYLCV incidence was interviewed (94%) regarded tomato as very low (0%-9%), could not recognize the most profitable of the horticultural whiteflies or TYLCV nor did they have crops they could grow and most (74%) local names for them. have cultivated tomato for more than 5 years. The varieties of tomato most All producers who recognized widely grown in Sudan were whiteflies and TYLCV had names for Moneymaker, Strain B, Peto 86, both and in most cases the producers Pearson, Early pack and Ace. Most gave very specific names for the producers (65%) bought their tomato problems they cause. Local names for seed from the local market, while 28% whiteflies include asala, biadah, used their own seed and 7% imported dubbana, dubbana beida, zubaba and seed from other countries. Other zubaba beida. These literally mean commonly grown vegetable crops “flies” or “white flies”, except asala, include cucurbits (especially melon and which refers to the honeydew that the cucumber), okra, common bean, insects secrete. Names given to the pepper, eggplant (Solanum melongena TYLC include hurug (burning), saratan L.), sweetpotato (Ipomoea batatas [L.] (cancer) and karmata, karmasha or Lam.), radish (Raphanus sativus L.), kurmut (all of which allude to leaf onion (Allium cepa L.), carrot (Daucus deformation).

122 Sudan

Historically, Sudan has been TYLCV problems every year. The regarded as one of the countries worst majority of producers (86%) believe hit by whitefly problems, especially in that there is a direct relationship the cotton production system. This between the climate and the incidence scenario has changed recently and the of whiteflies and/or whitefly- whitefly problem in cotton seems to transmitted disease, with 63% of have become less acute. Highly producers associating high incidence effective insecticide treatments, often with hot and dry seasons, while only containing insect growth regulators, 9% believe that they have the problem have reduced the breeding success of all year round. whiteflies on cotton. Nonetheless, the whitefly has maintained its status as a Estimation of disease highly injurious pest of tomato, incidence and yield losses cucurbits, pepper, okra and common The damage associated with B. tabaci bean, and the numbers of producers and WTVs is always more serious in who have abandoned tomato summer crops, especially during the cultivation because of TYLCV are on hot season from March to August the increase. (during which there are three planting periods: “early summer” in March, “late Farmers’ perceived yield loss to summer” in May and the “main whitefly and/or TYLCV in tomato was planting season” in July) and total on average 62%. According to the failure of the tomato crop due to TYLCV survey, 26% of producers, mainly in is common. This is despite the lower Gezira, reported total yield loss, 38% prevalence of the vector in the summer reported losing three-quarters of their season. Whitefly populations build up yield, 10% reported losing half, 10% during the winter (October-January) reported losing one-quarter of their and peak in December-January. Up to yield and 16%, mainly from Gedaref 100% infection by TYLCV can occur and North Kordufan, reported no loss. during a mild winter but less yield Of the producers interviewed, 34% depression is associated with these (most of them in Gezira State) reported winter infections. abandoning their tomato crops in at least 1 year because of whitefly/TYLVC In 1998, tomato grown during the problems and most reported that this May-July season suffered severe attack occurred in 1997. The perceived losses by TYLCV and this crop failed can be compared usefully with the completely. Watermelon likewise recorded incidence of TYLCV suffered severe losses from WmCSV. symptoms. Among all surveyed tomato An exceptionally heavy rainy season producers, 92% had some virus from mid-August to mid-October, symptoms in their tomato crop and however, seemed to alter the whitefly 50% had more than 25% incidence. It situation later in the year, with should be emphasized that 31% of the population levels dropping sharply and tomato producers had an 80% TYLC tomato crops grown during October- incidence in their tomato crops and November being comparatively less these outbreaks were mainly confined affected by TYLCV. A field experiment to the greater Gezira. TYLC was conducted at the University of Gezira significantly less severe (1%-9%) in revealed that early infection by TYLCV western, northern and eastern states, reduced the total number of flowers per and absent in Gedaref and the Abu plant, percent fruit setting and the Habil Basin. Most tomato producers quality of fruits. The infection resulted (65%) believe that they have whitefly/ in 67% reduction in the number of

123 Whiteflies and Whitefly-borne Viruses in the Tropics mature fruits and 84% reduction in the for seed production and water storage. weight of marketable fruits per plant. WmCSV incidence of 82% was observed Severe infection resulted in 86% on Galia melon grown for commercial reduction in the number of green purposes at the Gezira University farm mature fruits, 87% reduction in the in 1997. Complete absence of “netting” number of ripe fruits and 93% on the fruit surface was evident in some reduction in the weight of marketable severely affected plants, while the fruits per plant. Flower bud initiation healthy ones produced fruits entirely substantially decreased as the infection covered with netting (an important became more severe or when infection quality that correlates positively with occurred early in the plant’s growth sweetness of the fruit). The severely cycle but the proportion of flower affected plants gave 1.3 fruits per plant, abortion and fruit setting was not while healthy plants gave 2.1 fruits per significantly affected. Affected plants plant. The average incidence of WmCSV had reduced foliage cover and fruits in Galia melon from eight sites in the were exposed to direct sunlight, which Selait Irrigation Scheme (Kordufan caused white blotches (sunscald) and Province) was 70% and data from these reduced their market value. The results sites indicated a positive correlation show that TYLCV infection can affect between the level of virus incidence and all yield components, from flower yield loss. production through to the quantity and quality of fruits. OLCV restricts okra production in Sudan. Symptoms of the virus appeared to be more severe during the late winter WmCSV is less important and early summer growing seasons, economically than TYLCV but is with Gezira being the worst hit area. considered to be the most important The virus had less effect on local single cause of high yield loss of varieties of okra. BMMV may constitute cucurbits in eastern and central Sudan a potential hazard, especially in the and has affected the country’s melon newly established export-oriented exports seriously in recent years. Crops vegetable production area in Khartoum affected include watermelon, musk State. This virus is less important and is melon and, to a lesser extent, snake considered a late season disease in melon (Cucumis melo L. subsp. melo northern Sudan. var. flexuosus [L.] Naudin) and squash. Melon producers incur a 64% loss in revenue because of WmCSV. The virus Pesticide use is prevalent in Gezira, Upper Atbara The cost of controlling whiteflies and River and southern Sinnar, where fresh WTVs accounts for more than 30% of market varieties such as Charleston the total production costs of tomato in Grey, Congo Red and Sugar Baby are Sudan. A small proportion of grown. Watermelon in large areas of producers (4%) spent as much as Southern Blue Nile, Shuwak and US$300-400 per hectare on pesticide Dinder inland delta were totally treatment. Other producers estimated devastated by WmCSV in1998-99, and their pest control costs at US$200-299 up to 80% yield losses occurred in per hectare (14% of producers), Galia melon (a variety of musk melon) US$100-199 per hectare (21% of grown for export in Khartoum and producers), US$50-99 per hectare (27% Gezira States. However, this virus is of producers) and US$0-49 per hectare only sporadically important in (34% of producers). The costs of Kordufan and western Sudan where chemicals and their application vary local strains of watermelon are grown greatly among sites.

124 Sudan

Only 10% of producers reported tomato farming. A significant receiving technical advice or proportion of producers (30%) made information regarding the management more than 10 insecticide applications of pest and disease problems. Forty per season, 26% made 9 or percent of producers received advice 10 applications, and a few producers from other vegetable producers or applied pesticides as many as 24 to neighbours, 30% from commercial 30 times per season. Ten percent of sales agents, while 20% had not producers practiced crop rotation to received any advice but applied combat the problem. A few farmers insecticides according to their own used cultural control methods such as judgement. The absence of an effective burying of infected plants and inter- extension service has left the farmers cropping with repellent crops such as to make their own decisions on coriander (Coriandrum sativum L.) and whitefly/WTV management. The fenugreek (Trigonella foenum-graecum constraints on farmers’ knowledge of L.). Others claimed that directly sown whitefly management can be tomato, that is, without transplanting, summarized as: was less affected by TYLCV. Ten percent of producers, mainly in (1) Low level of education and limited Gedaref, Western State and Northern knowledge of modern techniques of State did not practice any sort of vegetable production; management of the problem.

(2) Poor ability to choose the right crop Lack of knowledge of the proper and a suitable variety of that crop; use of pesticides was evident in all areas surveyed. Farmers were found to (3) Poor knowledge on pesticide usage be ignorant of the judicious use of and related hazards; and pesticides and alternative whitefly (4) Inability to monitor market trends control methods. Some tomato growers as a basis for deciding what to overused chemicals, while others did grow and when to grow it. not spray until symptoms of damage were observed, and subsequent Because of the lack of extension intensive chemical application failed to services in vegetable farming, chemical save the crop. Producers were ignorant control of the whitefly remains the of the interval required between the principal pest management strategy. last spraying and harvest, and some The great majority of tomato producers even used the taste or smell of the in the country (80%) used insecticides, pesticides to judge their effectiveness. mostly pyrethroids and Lack of protective clothing and organophosphates, to combat the equipment exposed the farmers to whitefly/TYLCV problem on their serious health hazards. Inappropriate farms. The most popular pesticides chemicals such as fungicides applied by the producers were (propiconzale) and others without any fenpropathrin, fenvalerate and labels were being used in an attempt to cypermethrin (pyrethroids), omethoate, control the whitefly and WTV problem. malathion, dimethoate and Recommended dosages were not chlorpyriphos (organophosphates), followed and virtually all vegetable methomyl and carbaryl (carbamates), producers mixed one tomato sauce tin, and endosulfan (organochlorine). containing about 50 mL, of any insecticide, irrespective of its Most producers believed that the formulation or active ingredient, with pesticides were indispensable in 4 gallons (18 L) of water. A few

125 Whiteflies and Whitefly-borne Viruses in the Tropics producers used twigs to apply pesticide, whiteflies in the mixed vegetable instead of a knapsack sprayer, resulting cropping systems in the Sudan has not in poor coverage and greater loss of been documented. pesticides. Most producers (67%) applied insecticides as a preventive Sudan is the only country in East measure, 17% applied insecticides Africa that has adopted IPM as its when they observed whitefly/TYLCV official crop protection policy. The damage, while 8% applied insecticides government’s position is well defined in according to calendar. a recent publication entitled “Sudan Country Strategy Note, 1997-2001: Vegetable growers in Gezira and Partnership towards Sustainable Rahad illegally obtain part of their Human Development”. The document supply of chemicals from the stocks of emphasizes that the government of the cotton schemes. These include highly Sudan is pursuing an integrated toxic “cocktails” of insecticide, program for environmentally exclusively recommended for use on sustainable development. However, cotton. Their experience of severe implementation of the policy is minimal, whitefly outbreaks in previous seasons particularly, in the case of vegetable leads most of them to use excessive farming, because of farmers’ ignorance quantities of these chemicals. As a of alternatives to chemical control, result, environmental contamination, which in turn is mainly attributed to the occupational health hazards and lack of effective extension services in production costs are on the increase. In this sector. In the absence of addition, the overuse of chemicals to alternatives, chemical control of the control whitefly seemed to have resulted whitefly therefore remains the principal in the build-up of previously secondary strategy for managing WTVs. pests such as H. armigera, L. sativae and Scrobipalpa sp. Nevertheless, several producers suffered complete Strengthened Research loss of even the treated crop. Capacity

Several researchers (e.g., Ahmed et Sudan’s participation in the diagnostic al., 1987; Dittrich et al., 1990) have phase of the TWF-IPM Project has documented the resistance status of provided a substantial body of B. tabaci in cotton to various knowledge covering various aspects, insecticides in the Sudan. They biological and socio-economic, of the reported that B. tabaci showed high whitefly/WTV problems in vegetable- levels of resistance to many based cropping systems in the country. organophosphates and synthetic Sudan’s participation in the planning pyrethroids and moderate resistance to workshop in April 1997 and the carbamates and organochlorines. The co-ordination meeting in June 1998 specific levels of resistance reported provided national researchers with a vary depending on the source of the chance to learn from, and share ideas samples, the history of insecticide and experiences with, scientists from application in that area, the time of the other countries. A Ph.D. project on year when bioassays were done and the “Distribution of whitefly species (and sensitivity of the whitefly population biotypes) and their natural enemies in used. However, the reports were vegetable-based cropping systems in primarily based on dose responses of Sudan” was undertaken in collaboration whiteflies in the cotton monoculture with the University of Gezira under the system. Pesticide resistance of auspices of the project.

126 Sudan

Conclusions The surveys documented alarming misuse of pesticides on vegetables. The The work conducted during the information gained can help to diagnostic phase of the TWF-IPM sensitize policy makers in Sudan to the Project has provided, for the first time, alarming magnitude of the whitefly and a biological and socio-economic WTV problems and to the dangers characterization of the problems linked inherent in the associated misuse of to B. tabaci in the vegetable cropping pesticides. The knowledge gained has systems of Sudan. The studies have provided a sound foundation for provided a better understanding of the developing a future research agenda. biology of this important vector, the The following priorities are distribution of its natural enemies, the recommended: epidemiology of the viruses it transmits and the socio-economic implications of (1) Research should be conducted on virus disease problems for producers. the following IPM components:

Farmers viewed the B. tabaci/ (a) Evaluation of available TYLCV problem as the most important TYLCV-resistant or -tolerant constraint on tomato production. The tomato varieties for immediate areas identified as hot spots for TYLCV incorporation into IPM plans; include Gezira, Managil, Umelgura and Hasaheisa Provinces, all of which are (b) Screening of exotic tomato characterized by irrigated production varieties for resistance to and are within cotton production areas. B. tabaci and/or TYLCV; The main problems associated with (c) Breeding for resistance against B. tabaci in vegetable production B. tabaci and WTVs, while systems in Sudan are: utilizing resistant strains of crops; (1) Tomato: transmission of TYLCV and TVTV, and direct feeding by (d) Evaluation of alternative and B. tabaci during early crop stages; novel pest and disease management methods, (2) Watermelon and musk melon: including the use of botanicals, transmission of WmCSV, CVYV, biological control and cultural and a virus inducing potyvirus-like methods; and yellowing symptoms; (e) Investigation of the efficiency of (3) Okra: transmission of OLCV, sooty natural enemies in vector mould induced by secretion of control. honeydew by B. tabaci, and direct feeding damage; (2) Available IPM options should be tested and refined, with the close (4) Peppers: transmission of TYLCV; participation of producers, under and on-farm conditions; participatory (5) Vegetable legumes: symptoms of a research should be complemented begomovirus and closterovirus with intensive training of producers observed on cowpea and common and other stakeholders. bean but transmission by B. tabaci (3) Resistance of whiteflies to is not yet confirmed. insecticides commonly used in vegetable farming should be

127 Whiteflies and Whitefly-borne Viruses in the Tropics

monitored, using a standardized Dittrich, V.; Ruesch, O.; Ernest, G. H. methodology in Sudan and 1990. Resistance mechanism in the elsewhere to enable valid sweetpotato whitefly (Homoptera: Aleyrodidae) populations from comparisons to be made. Sudan, Turkey, Guatemala and Nicaragua. J. Econ. Entomol. (4) Appropriate techniques to evaluate 83:1665-1670. insecticide residues in fruits and vegetables should be developed and Eveleens, K. G. 1983. Chemical insect applied. control in the Sudan Gezira: Analysis of a crisis crop. Crop Prot. (5) The whiteflies observed inducing 2:273-287. silverleaf in Gezira for the first time in 1998 should be characterized Kirkpatrick, T. W. 1931. Further studies on leaf-curl of cotton in the Sudan. using molecular techniques and Bull. Entomol. Res. 22:323-363. biological assays. Shapiro, J. P. 1995. Insect-plant The recommended research will interactions and expression of provide a platform for the development disorders induced by the silverleaf and adoption of sustainable production whitefly, Bemisia argentifolii. In: practices for vegetables in Sudan, Gerling, D.; Mayer, R. T. (eds.). leading to improved well-being for Bemisia: Taxonomy, biology, producers and consumers. damage, control and management. Intercept, Andover, GB. p. 167-171.

Yassin, A. M.; Nour, M. A. 1965. Tomato References leaf curl disease, its effects on yield and varietal susceptibility. Sudan Ahmed, A. H. M.; Elhang, E. A.; Bashir, Agric. J. 1:3-7. H. H. 1987. Insecticide resistance in the cotton whitefly (Bemisia tabaci Genn.) in the Sudan Gezira. Trop. Pest Manag. 33:67.

Dafalla, G. A.; Sidig, S. A. 1997. Management of the whitefly-virus complex in Sudan. Food and Agriculture Organization (FAO) Plant Production and Protection Paper 143. Rome, IT. 214 p.

128 Kenya

CHAPTER 2.3 Kenya

Mohamed Ali Bob*, Benjamin Odhiambo**, Gilbert Kibata** and Jason Ong’aro**

Introduction

Geographical context Eastern Field surveys covering 94 farms were conducted in 1997-98 following the Kenya methodology agreed among project Kitale partners. The farms represented: Western Kirinyaga, Thika, Muranga, Kiambu, Northeastern and Nairobi Districts in Central Kisumu Central Province; Machakos, Athi River, Meru, Nakuru Kitui, and Kibwezi Districts in Eastern Province; Kwale, Kilifi, and Taita Taveta Nairobi Districts in Coast Province; Kajiado and Naivasha Districts in Rift Valley Towns Province; Kisii, Migori, Homa Bay, and Surveyed Siaya Districts in Nyanza Province; and areas Vihiga and Busia Districts in Western Province (Figure 1). The districts were selected based on the prevalence of the Figure 1. Tomato-growing areas surveyed for whitefly incidence in Kenya. cultivation of tomato (Lycopersicon esculentum Mill.). Table 1 gives the mean annual temperatures and Table 1. Mean annual temperatures and precipitation for each of the surveyed precipitation for the surveyed regions of Kenya. regions. Region/ Mean min.-max. Mean Province temperature precipitation In the surveyed regions, tomato is (°C) (mm) planted throughout the year with the Eastern 15-26 728 exceptions noted below. Eastern Rift Valley 14-27 591 Province is a region of semi-arid lower Central 13-25 1166 mid-altitude land becoming transitional Coast 21-30 872 semi-humid towards Central Province; Western and 16-29 1289 here, tomato is not planted in July. Nyanza

* International Center of Insect Physiology and Whitefly surveys took place between Ecology (ICIPE), Nairobi, Kenya. ** Kenya Agricultural Research Institute (KARI), November and March of 1997-98 and Nairobi, Kenya. 1998-99. Rift Valley Province consists

129 Whiteflies and Whitefly-borne Viruses in the Tropics of transitional lower to higher mid- Cucurbitaceae and Euphorbiaceae, altitude land. In the surveyed areas of particularly in the semi-arid Eastern Kajiado and Naivasha Districts in the Province of Kenya. Extension workers Rift Valley, tomato is not planted in and scientists have reported January and July. Whitefly surveys informally the occurrence of several were carried out in September 1998 whitefly-transmitted viruses: Tomato and March 1999. Central Province yellow leaf curl virus (TYLCV), Cowpea consists of sub-humid highlands; mild mottle virus (CPMMV), whitefly surveys were carried out here Sweetpotato mild mottle virus (SPMMV) between November and March in and Watermelon chlorotic stunt virus 1997-98 and in 1998-99. In the Coast (WmCSV). Province, a region of sub-humid coastal lowlands, tomato is not Despite the perceived importance planted from November to January. of these horticultural crops and their Whitefly surveys were carried out in virus disease problems in Kenya, July 1998. In Western and Nyanza previously no systematic attempt has Provinces (sub-humid to humid upper been made to document their mid-altitude land and lower highlands) distribution and incidence. Virtually tomato is not planted in January. all work related to whiteflies Whitefly surveys in this region were conducted in Kenya has focused on carried out in February and November cassava mosaic disease (CMD), its 1998. prevalence in the country, its temporal and spatial distribution and El Niño rains devastated tomato associated yield loss (Seif, 1982; Bock, production in Kenya from September 1987). Some research also had been 1997 to June 1998. Whiteflies were conducted on the citrus woolly scarce during this period. whitefly Aleurothrixus floccosus (Maskell), which is reported to have The emergence of Bemisia become a common pest of citrus tabaci as a pest and virus (Citrus spp. L.) as well as coffee vector (Coffea arabica L.) and guava (Psidium Tomato is one of the most widely guajava L.) in Kenya (Löhr, 1997). In cultivated vegetable crops in East addition, the whitefly parasitoids, Africa, predominantly grown by small- Eretmocerus mundus Mercet. and scale producers for fresh consumption Encarsia sublutea (Silvestri), have (GTZ, 1995), while in Kenya in been recorded in Kenya (Dr. Abdurabi particular, vegetable-growing both for Seif, personal communication, 1998). local consumption and for export Kenya’s participation in the Tropical provides an increasingly important Whitefly Integrated Pest Management source of income to farmers. In 1995, (TWF-IPM) Project provided an Kenyan national scientists opportunity to gain an overview of the participating in a regional workshop geographical range and economic ranked the whitefly Bemisia tabaci importance of whiteflies, whitefly- (Gennadius) as the most important transmitted viruses and whitefly pest/vector in tomato (Varela and natural enemies in the country— Pekke, 1995). This whitefly has information that would serve as a become an important production sound basis for planning further constraint to crops of the families research and pest management Solanaceae, Leguminosae, Malvaceae, efforts.

130 Kenya

Increased Biological Solanaceae and Verbenaceae (Table 3). Understanding Surprisingly, whiteflies were found breeding only occasionally on tomato in Kenya. In Muranga, Kiambu and Mwea Characterization of Districts in Central Province, it was begomoviruses and whitefly difficult to find a single whitefly nymph biotypes in a whole tomato field. In these areas, During the surveys in Kenya, whiteflies, mainly T. vaporariorum, were 325 collections of whitefly adults and found breeding on crops such as nymphs were made. From these, common bean (Phaseolus vulgaris L.) 549 specimens of whitefly nymphs and cassava and on non-cultivated were mounted on 310 slides and plants, most significantly Euphorbia identified to species level. Most of the heterophylla L., in all areas surveyed. whiteflies collected were B. tabaci However, in Kihara (Nairobi) and (57%), while Trialeurodes vaporariorum Kigumo (Muranga) in Central Province (Westwood) represented 34%. B. afer and in Nguruman in Rift Valley (Priesner and Hosney) were Province, whiteflies, mainly occasionally found (7%) in fields of T. vaporariorum, were found breeding cassava (Manihot esculenta Crantz) readily on tomato. adjacent to tomato fields. A number of other whitefly species (2%) also were Host plant feeding preference and encountered, including Trialeurodes status as reproductive hosts ricini (Misra), A. floccosus, Siphoninus (oviposition and offspring survival) of phillyreae (Haliday) and Bemisia spp. B. tabaci was studied in field cages at Table 2 shows the composition of the International Center of Insect whitefly species based on the identified Physiology and Ecology (ICIPE), specimens from each Province. Nairobi, in 1997 using adult insects collected from a tomato field nearby. Disease incidence and Among the four host plants studied, symptom severity common bean was the most preferred host plant, followed by lablab (Lablab Whiteflies in the surveyed farms were purpureus [L.] Sweet), cowpea (Vigna found breeding on host plants that unguiculata [L.] Walp.) and okra belong to the following families: (Abelmoschus esculentus [L.] Moench). Acanthaceae, Amaranthaceae, There were significant differences Asteraceae, Commelinaceae, between plants with regard to the Convolvulaceae, Cucurbitaceae, parameters tested. These results Euphorbiaceae, Lamiaceae, concur with observations in the field Leguminosae, Malvaceae, Rutaceae, during the survey.

Table 2. Species composition (%) of whitefly samples from surveyed regions of Kenya.

Province Species compositiona

B.t. T.v. B.a. T.r. A.f. S.p. Eastern 61 29 6 0 0 4 Rift Valley 20 80 0 0 0 0 Central 35 60 4 1 0 0 Coast 79 0 21 0 0 0 Western 90 0 7 3 0 0 Nyanza 79 6 5 0 10 0 a. B.t., Bemisia tabaci; T.v., Trialeurodes vaporariorum; B.a., Bemisia afer; T.r., Trialeurodes ricini; A.f., Aleurothrixus floccosus; S.p., Siphoninus phillyreae.

131 Whiteflies and Whitefly-borne Viruses in the Tropics Guerke L. (L). Griseb. (Oliv.) Wild L. L. Benth. ex Oliv. L. (L.) Gaertn. Cav. L. T. Anders. (Hemsl.) A. Gray (Hemsl.) A. Gray (Hemsl.) A. Gray (L.) R. Br. (L.) R. Br. L. L. (Andrews) Meikle & Verdc. L. (Oliv.) C. Jeffrey Cav. L. Burm f. Unidentified tree Leonotis nepetifolia Tithonia diversifolia Euphorbia heterophylla Datura stramonium Lantana camara Tithonia diversifolia Nicandra physalodes Sida acuta Ocimum kilimandsharicum Ipomoea tricolor Jacquemontia tamnifolia Rhynchosia hirta Leonotis nepetifolia Bidens pilosa Galinsoga parviflora Gutenbergia cordifolia Tithonia diversifolia Commelina benghalensis Euphorbia heterophylla Aspilia mossambicensis Emilia discifolia Achyranthes aspera Achyranthes sicula Ageratum conyzoides Asystasia schimperi Asteraceae Euphorbiaceae Solanaceae Verbenaceae Asteraceae Malvaceae Mill. Lamiaceae Mill. Mill. Lamiaceae (L.) Moench (L.) Moench L. L. L. Euphorbiaceae L. L. L. L. L. Commelinaceae L. Leguminosae L. Convolvulacea Crantz (L.) Walp. (L.) Sweet (Thunb.) Matsum & Nakai Amaranthaceae L. Lam. (L.) Lam. Acanthaceae antia L. Asteraceae L. sp. sp. spp. spp. Phaseolus vulgaris Gossypium hirsutum Citrus Ipomoea batatas Lycopersicon esculentum Phaseolus vulgaris Abelmoschus esculentus Lycopersicon esculentum Solanum melongela Momordica charantia Cucurbita Abelmoschus esculentus Solanum melongena Ocimum basilicum Gossypium hirsutum Lycopersicon esculentum Capsicum annuum Crotalaria Lablab purpureus Crotalaria Vigna unguiculata Phaseolus vulgaris Manihot esculenta Ipomoea batatas Citrullus lanatus Cucurbita pepo Momordica char Leguminosae Malvaceae Convolvulaceae Solanaceae Leguminosae Malvaceae Solanaceae Malvaceae Lamiaceae Solanaceae Leguminosae Euphorbiaceae Convolvulaceae Cucurbitaceae Family Species Family Species Hosney) (Westwood) Cucurbitaceae (Maskell) Rutaceae (Haliday) (Misra) (Gennadius) (Priesner and Siphoninus phillyreae Aleurothrixus floccosus Trialeurodes ricini Bemisia afer Trialeurodes vaporariorum Bemisia tabaci Whitefly species Crops Non-cultivated hosts Table 3. Reproductive host plants of whitefly species encountered during surveys in Kenya.

132 Kenya

Samples of 10 plants per field, on of TYLCV. Severe begomovirus average, collected from all surveyed symptoms also were observed on a areas, were squashed on nylon weed identified as Gutenbergia membranes and sent to the John Innes cordifolia Benth. ex Oliv. E. Centre, UK, for hybridization. heterophylla was found to be a major Hybridization signals varied from reproductive host of both B. tabaci and strong to very weak, probably because T. vaporariorum at any site where of differences in the degree of nucleic whitefly infestation occurs and this acid homology. This could suggest plant species is present. Therefore, the different strains of begomoviruses or role of E. heterophylla in survival of the different begomoviruses altogether. The disease and the vector outside the signal strength also may have been crop-growing season needs to be influenced by the sampling method, investigated. because the virus titre may have been low in the particular tissues sampled or Natural enemy species at that stage of infection. Forty whitefly parasitoid samples were obtained from the surveys and At Kibwezi, begomovirus symptoms preliminary identification was carried were observed on sweet pepper out. Encarsia sophia (Girault and Dodd) (Capsicum annuum L.) and a low was found to be the predominant incidence of WmCSV (about 5%) was species, while Eretmocerus sp. was recorded on watermelon (Citrullus encountered in the Central Province. lanatus [Thunb.] Matsum. & Nakai). A Encarsia formosa Gahan was only variety of mosaic symptoms that need collected from Mwea in the Central further investigation also were noticed Province. At Nguruman, Rift Valley, the in pepper and watermelon. There were level of parasitization appeared higher symptoms that suggest that tomato in than at other survey sites in Kenya. this region should be tested for potyviruses such as Potato virus Y (PVY). Aphid infestation levels were Some of the 24 predators collected very high at Kibwezi and some of the were found feeding on the whitefly symptoms observed could be a result of pupae. Most were coccinnellids, while a aphid-borne viruses. High incidence few were spiders and mites but none and severe symptoms of CMD were were identified to species level. observed in Mombasa and Kilifi Macrolophus caliginosus Wagner, bugs Districts (Coast Province) and in Kitui known to prey on T. vaporariorum in District (Eastern Province). The other parts of the world, were collected symptoms appeared more severe than at several locations in Nguruman (Rift those observed in western Kenya. Valley Province), Kibwezi (Eastern Whether these symptoms are caused province) and Mwea (Central Province). by infections of East African cassava mosaic virus (EACMV) or are associated Geographical range of whitefly with the occurrence of the Ugandan and whitefly-transmitted virus variant needs to be ascertained. infestation Areas identified as “hot spots” for Leaf curl symptoms were found TYLCV infection include Kibwezi, Kitui, during the survey on non-cultivated Machakos, Athi River (Eastern host plants. These were Achyranthes Province) and Naivasha (Rift Valley aspera L., E. heterophylla and Nicandra Province). Low incidence of TYLCV physalodes (L.) Gaertn. These plants symptoms was observed in all other are therefore likely to act as reservoirs surveyed provinces with the exception

133 Whiteflies and Whitefly-borne Viruses in the Tropics of Western and Nyanza Provinces, from December to early March (hot and where no TYLCV symptoms were dry season) but the whiteflies virtually observed. disappear from April to August (long rainy season followed by cold season), B. tabaci was found to be the most despite the presence of the same crops common whitefly species in the country in the irrigated fields throughout the (Table 3). T. vaporariorum was common year. in the highlands and was encountered frequently in the Central, Rift Valley High T. vaporariorum populations and Eastern Provinces. T. vaporariorum were observed in Nguruman (Rift Valley was not recorded from the Coastal and Province) where the underside of leaves Western Provinces. The abundance of tomato, common bean and eggplant pattern and economic importance of (Solanum melongena L.) could be found B. tabaci/TYLCV differed significantly covered with adults and nymphs. In between provinces. It was found to this area, the infestation level only attain particularly high epidemic drops during a brief rainy season, incidence (>25% of the tomato with around April-May. A similar level of TYLCV symptoms) in Eastern Province infestation by the same whitefly (Kibwezi 30%-100%, Kitui 30%-100%, species was observed in the relatively Machakos and Athi River 15%-30%) cooler area of Kihara (Central and in Rift Valley Province (Naivasha Province). Direct feeding damage by 15%-40%), where whiteflies and TYLCV T. vaporariorum was observed in both were of great concern to the tomato Central Province (Kihara and Mwea) producers. This suggests that the and Rift Valley Province (Nguruman), impact of the B. tabaci/TYLCV complex where it was the predominant whitefly is generally more severe in the dry species. areas. Since completion of the survey, In Coast Province, where the very high whitefly infestations have weather is hot and humid throughout been reported in the cut-flower and the year, B. tabaci constituted 79% of ornamental plants on export-oriented the collected whitefly specimens. flower and vegetable farms around However, only a moderate incidence of Lake Naivasha (Rift Valley Province). TYLCV symptoms (5%-25%) was The highest TYLCV incidences were observed. In Western Province, observed at altitudes ranging from B. tabaci constituted 90% of the about 500 to 1000 m altitude (Table 4). collected whitefly specimens and in B. tabaci was the dominant whitefly Nyanza Province, 79% (weather is also species up to an altitude of about hot and humid throughout the year) 1700 m, above which T. vaporariorum but no TYLCV symptoms were observed became dominant. on tomato, other vegetable crops, or weeds, and collected samples were Increased Socio-economic negative when hybridized using a probe Understanding to the original Israeli isolate of TYLCV. In Western and Nyanza Provinces, most producers plant tomato as a secondary Farmers’ assessment of crop after the main staple cereal crops, whitefly-related problems maize (Zea mays L.) and sorghum Smallholders dominate vegetable (Sorghum bicolor [L.] Moench). The production in Kenya and account for B. tabaci population at Kibwezi 70% of the total production (Mulandi, (Eastern Province) is usually very high 1998). Producers in the survey areas

134 Kenya

Table 4. Maximum incidence (% of sampled plants showing disease symptoms) of Tomato yellow leaf curl virus (TYLCV) encountered and species composition (%) of whitefly samples collected at different altitudes in Kenya.

Altitudea (m) TYLCV Whitefly species compositionb incidence B.t. T.v. T.r. B.a. S.p. A.f. B. spp.

20-270 8 72 6 0 14 8 0 0 520-770 100 81 6 0 13 0 0 0 770-1020 80 33 65 2 0 0 0 0 1020-1270 30 66 25 1 8 0 0 0 1270-1520 15 81 16 0 1 0 0 2 1520-1770 27 77 9 2 5 0 7 0 1770-2020 39 43 53 0 4 0 0 0 a. No survey was conducted in the 270-520 altitude range. b. B.t., Bemisia tabaci; T.v., Trialeurodes vaporariorum; T.r., Trialeurodes ricini; B.a., Bemisia afer; S.p., Siphoninus phillyreae; A.f., Aleurothrixus floccosus; B. spp., Bemisia spp. were surveyed using agreed protocols. cepa L.). Most producers (94%) Most of the interviewed producers practised crop rotation. (75%) used their own land, while 13% used rented lands and 12% had other Fifty-five percent of the interviewed arrangements. The great majority of producers listed whiteflies and/or producers (80%) were men. associated virus diseases among their principal pests and diseases and Tomato is one of the most ranked these problems, on average, as important vegetable crops in the their third-most-important problem in country. Virtually all interviewed tomato. For tomato producers in producers (98%) regarded tomato as Kenya, late blight Phytophthora their most profitable vegetable crop and infestans (Mont.) de Bary is the main 58% of the producers interviewed have cause of concern as a yield-limiting cultivated tomato for more than factor. Other pest and disease 5 years. Main varieties of tomato grown problems were early blight (Alternaria in the country were Moneymaker and solani [Ell. and Mart.] Jones and Cal-J, while other varieties include Grout), red spider mites (Tetranychus Roma, Marglobe, Petomech, Early urticae Koch.), bollworms (Helicoverpa Beauty, Mecheast, Bonny Best, Heinz, armigera [Hubner]), leaf miners Floradade, Marmande, Ponderosa and (Lyriomyza spp.), bacterial wilt (Erwinia Hotset. However, no tomato variety tracheiphila [Smith]), aphids, cutworms grown in Kenya was known to be (Spodoptera litura [Fabricius]), resistant or tolerant to TYLC. Most nematodes (Rhabditida: producers (75%) bought their tomato Steinernematidae & Heterorhabditae), seed from the local market, while 20% end rot (Godronia cassandrae f. used their own seed and 5% reported vaccinii), thrips (Thrips palmi Karny) buying directly from the seed and rats (Rattus sp.). companies. Other vegetable crops include kale (Brassica oleracea L.), Most producers (88%) were able to common bean, sweet pepper, cabbage recognize whiteflies, while only 64% of (Brassica oleracea var. capitata L.), producers were able to recognize the eggplant, sweetpotato (Ipomoea batatas symptoms of TYLCV and only 1% knew [L.] Lam.), okra, bitter gourd that whiteflies and TYLCV were inter- (Momordica charantia L.), melon related. Although only about half of the (Cucumis melo L.) and onion (Allium producers (55%) listed whiteflies and

135 Whiteflies and Whitefly-borne Viruses in the Tropics associated viruses among their Estimation of disease principal pests and diseases, 79% of incidence and yield losses interviewed producers believed that Among the surveyed farms, 55% had whiteflies cause problems in the crops TYLCV symptoms in their tomato crop. on their farms. Among these However, very low incidences (1% producers, 50% attributed the infection) were found in most parts of problems to both whiteflies and TYLCV, Central, Eastern and Coastal 47% attributed the problem to Provinces. Incidence of TYLCV whiteflies only, while only 3% symptoms of more than 30% was mentioned TYLCV alone. recorded in 11 out of 94 farms. These were in Kibwezi (3), Kitui (3), Areas where producers were not Machakos/Athi River (1) and Naivasha able to recognize whiteflies were mainly (4). High incidence (80%-100%) was in Western Province (Funyula, Mokonja only recorded in two farms in Eastern and Buhuma), Nyanza Province (South Province, in Kibwezi and Kitui areas. Ugenya and Nyabondo) and Coast The incidence in Kibwezi resulted in a Province (Shimba Hills and Kikambala), total loss of the tomato crop. Severe where whiteflies were scarce. Most leaf curling plus subsequent yield producers in these localities also could depression of nearly 100% in sweet not recognize TYLCV symptoms and pepper was also observed in one were not aware of its association with experimental plot in Kibwezi. Despite the vector. Some thought that TYLCV the presence of B. tabaci in western was either a soil-borne or seed-borne Kenya (Western and Nyanza Provinces), problem, while some confused it with no symptoms of TYLCV infection were blight. recorded in this part of the country.

Most producers who recognized On average, producers’ estimate of whiteflies and TYLCV symptoms had the yield loss in tomato production was names for both but in most cases the 30%. According to the survey, 1% producers gave non-specific names for reported a total yield loss, 5% reported whiteflies (e.g., “insects”) and for three-quarter yield loss, 26% reported TYLCV (e.g., “disease”). Most of the half yield loss and 50% of the names given to the whiteflies literally producers reported one-quarter yield mean “flies” or “small insects” and local loss due to the whitefly/TYLCV bichuni names include e (insect), complex. The problem was more keguneta, kimbulutwa (butterfly), kudni serious in drier areas where irrigation mbuu ngaturia (insect), (mosquito), , was practised and this could be okogataa oulolo (mites), (white moulds), because of the presence of TYLCV rwagi rweru (white mosquito), twihuruta reservoirs and the virus vector. tweru ume (small white butterflies), Eighteen percent of the producers umuu vipuvute (aphids), (moulds), interviewed, mainly from Western vidogo and obororo (plant fleas). Names Province, thought that their tomato given to the disease problem mean “leaf crops did not suffer any yield loss curling”, “folding” or “stunted plant because of the whitefly/TYLCV growth” and local names include chana complex. However, 8% of producers matawi gathuri (combing of leaves), interviewed reported abandoning their gikware gukunja (stunted old man), , tomato crops in at least 1 year because mathangu (folding of leaves), kanyaria of the problem. These were from kugogonyara (the devastating disease), Eastern, Rift Valley, Central and Coast mbaa (curly) and (blight). Provinces. Half of them reported that the loss occurred in 1997.

136 Kenya

Seventy-nine percent of tomato routinely according to the calendar. producers believe that they encounter The high level of insecticide usage the whitefly/TYLCV complex every coupled with prophylactic chemical year. Almost one-third of the producers spraying could easily lead to the (30%) believed that the incidence and whiteflies developing resistance, which severity of whitefly-transmitted viruses, would hamper their control. especially in tomato, was exceptionally high in early 1997 before the El Niño phenomenon. Most (89%) also believe Because of the lack of virus- that there is a direct relationship resistant tomato cultivars, producers between the seasonal weather changes resort to the use of pesticides to and the whitefly/disease incidence. combat the whitefly/TYLCV complex. Seventy-five percent of producers The use of pesticides in most reported that the period from production areas in Kenya has December to March (hot and dry increased, even though it is hazardous season) was the time when they and sometimes ineffective and experience the most serious whitefly/ uneconomical. Eighty-six percent of TYLCV problem, while few believed that tomato producers used insecticides, they have the problem all year round. mostly pyrethroids and Producers gave estimation of the costs organophosphates, to combat the involved (see below). whitefly problem on their farms. However, 9% of the producers did not Costs estimated by producers practice any sort of control against the in control of whitefly/TYLCV problem, while 4% practised crop complex per hectare of tomato rotation as a means of combating it. Some producers used a Bacillus US$ % producers thuringiensis formulation (Dipel) for 0-49 13 control of whiteflies. One producer in 50-99 17 Mwea (Central Province) reported using 100-199 30 ash, in addition to commercial 200-299 17 chemicals, to manage the whitefly/ 300-400 23 TYLCV complex on his farm. Another producer in Kihara (Central Province) Pesticide use used a home-prepared blend of Almost half of the producers (43%) botanicals and a detergent to manage received advice and recommendations the whitefly problem but apparently on the use of insecticides from with little success. Chemical control technical advisors, 37% from other was in many cases directed against producers, neighbours or family blight (i.e., using fungicides). The most members, 10% from pesticides commonly used insecticides for control salesmen, while 10% claimed to use of the whitefly/TYLCV complex in the chemicals on their own initiative. Most country were pyrethroids (lambda- producers (85%) made their own cyhalothrin and cypermethrin) and decision on what, when and which organophosphates (dimethoate and insecticides to apply, while only 10% diazinon). Other commonly used relied on the technical advice they insecticides include alpha received. Half of the producers (48%) cypermethrin (pyrethroid), amitraz applied insecticides as a preventive (amidine), malathion, fenitrothion, and measure, 25% applied insecticides omethoate (organophosphates) and when they observed whitefly/TYLCV endosulfan (organochlorines). damage, while 18% applied insecticides Acaricides such as dicofol, which has

137 Whiteflies and Whitefly-borne Viruses in the Tropics little effect on insects (EXTOXNET, glass-vial and leaf-dip bioassays was 1998), were also commonly applied carried out at Kibwezi, Kitui (Eastern against whiteflies. The use of fungicides Province), Nguruman (Rift Valley such as metalaxyl, copper sulphate and Province) and Mwea and Kihara (Central Mancozeb (ethylene Province). Generally, T. vaporariorum bisdithiocarbamate), in an attempt to was found to be more resistant to the combat TYLCV was very common. three insecticides than was B. tabaci. Of Almost half the producers (48%) made the three insecticides used in the tests, between one and four insecticide cypermethrin was the most commonly applications per crop season to control used in the survey farms and bifenthrin the whitefly/TYLCV complex, while only the least. The resistance levels deduced 8% applied nine to ten applications. from the analysis showed that both B. Most producers in Nyanza and Western tabaci and T. vaporariorum were more Provinces did not apply insecticides or resistant to cypermethrin than to either practice any sort of control measure bifenthrin or methomyl. The results against the whitefly/TYLCV complex. obtained suggest that repeated exposure of whiteflies to specific There was lack of knowledge on insecticides could have selected them proper use of pesticides. Some tomato for resistance to the chemicals. growers overused chemicals, while However, this was not a universal others did not spray until symptoms of conclusion because there were some damage were observed, and subsequent deviations from the general trend. For intensive chemical application failed to instance, a 6.8-fold resistance to save the crop. In many cases, producers bifenthrin was detected at Kihara where continued spraying on a crop already the insecticide had reportedly never severely affected by the virus, using a been used. Whiteflies from Kajiado were higher dosage in the hope of curing the the most resistant to cypermethrin, plants. Producers were ignorant about giving a resistance factor value (RF) of the safety period between the last 87 in the leaf-dip method, while those spraying of the crop and date of sale or from Nairobi were the most resistant to consumption. Some producers said they methomyl. did not care about the pre-harvest interval as long as the produce looked good for sale. Some of the local Strengthened Research agrochemical suppliers bought large Capacity volumes of pesticides and measured out small volumes according to the financial The present work has substantially capability of the buyers. The application improved our understanding of the of inappropriate pesticides and dose whitefly problem in the major vegetable rates was common and led to control production areas of Kenya. Quantitative failure. Chemicals without any labels information on whiteflies and natural were usually bought from local dealers. enemy species diversity and the occurrence of the whitefly-transmitted For the first time in Kenya, diseases, mainly TYLCV, as well as the information was generated on whitefly socio-economic understanding of the resistance to insecticides in selected producers’ perception and practices areas of the country. Assessment of the relating to management of the whitefly/ resistance of B. tabaci and T. TYLCV complex, has provided Kenyan vaporariorum to methomyl (a researchers with a solid background for carbamate), cypermethrin and setting a research agenda for further bifenthrin (pyrethroids) using both work, some of which may be carried out

138 Kenya under Phase 2 of the TWF-IPM Project example, very heavy whitefly infestation (see Introduction to this volume). The was observed in two adjacent farms at first, diagnostic, phase of the project Kihara but not in any of the other also contributed to capacity building nearby farms. The contrast could have through a planning workshop and a been due to the insects’ feeding mid-term meeting for all participating preference, because the two farms had scientists, training of one M.Sc. thesis eggplant and squash (both preferred student in assessment of insecticide hosts) that were not on the other farms resistance and training of one B.Sc. but this possibility would need to be student in host plant preference investigated further. We need to studies. understand why there is low or no incidence of TLCV in western Kenya, despite the prevalence of the vector. Conclusions The potential of local isolates and commercial formulations of The following areas in Kenya have been entomopathogenic fungi, as well as identified as hot spots for future parasitoids and predators, for control research on whiteflies and whitefly- of whiteflies could be evaluated. There transmitted viruses: Kibwezi, Kitui, is also a need to look at other whitefly- Machakos/Athi River and Naivasha. transmitted viral diseases and at the The survey revealed that the prevalent role of weeds as reservoirs of whitefly species in the country are begomoviruses. B. tabaci, T. vaporariorum, B. afer, T. ricini, A. floccosus and S. phillyreae. There is a need for training of The fact that 55% of the surveyed entomologists and virologists in areas farms had TYLCV symptoms indicates of detection, identification and a potential for whitefly/TYLCV characterization of whitefly-transmitted problems in Kenya. The major viruses, whiteflies and their natural constraint in producers’ knowledge on enemies, as well as training in the whitefly/TYLCV complex was a epidemiological research complete lack of understanding (99% of methodologies. The continued farmers) of the connection between the involvement of scientists from ICIPE disease and the vector. and the Asian Vegetable Research and Development Center (AVRDC), who To minimize the devastating effects have provided technical support during of whitefly-transmitted viruses in the first phase of this project, would be tomato, sustainable vector- and helpful. Collaborative work involving disease-management options need to the network of regional stations of the be developed, tested and deployed, Kenya Agricultural Research Institute including the use of resistant and/or (KARI) should be extended. tolerant cultivars, as well as cultural and biological control practices. The development of such options needs to References be based on, and supported by, appropriate research. In addition, we Bock, K. R. 1987. Some aspects of the need to understand the source of epidemiology of African cassava whitefly infestations and the reasons mosaic virus in coastal districts of Kenya. In: Procs. International behind the striking between-field Seminar on African cassava mosaic variability in whitefly population levels disease and its control, 3-4 May that is often observed among fields in 1987, Yomoussokro, CI. close proximity to each other. For p. 125-130.

139 Whiteflies and Whitefly-borne Viruses in the Tropics

EXTOXNET (Extension Toxicology Mulandi, M. A. S. 1998. The status of the Network). 1998. Pesticide horticultural industry, challenges information profile, 3 Nov 1998. and strategies, now and beyond the Available in: http://pmep.cce. year 2000. Horticultural Crops cornell.edu/profiles/extoxnet/ Development Authority, Nairobi, carbaryl-dicrotophos/dicofol- KE. 34 p. ext.html. (Accessed 2 May 2000) Seif A. A. 1982. Effect of the cassava GTZ (Deutsche Gesellschaft für mosaic virus on yield of cassava. Technische Zusammenarbeit). Plant Dis. 66:661-662. 1995. Major pests of tomatoes in eastern and southern Africa: Varela, A. M.; Pekke, A. 1995. Compilation of past research work Proceedings of Tomato Planning and identification of IPM Workshop for Eastern and Southern opportunities. GTZ-Integrated Pest Africa Region, 16-20 October 1995, Management (IPM) Horticulture Harare, ZW. Deutsche Gesellschaft Project Report, Nairobi, KE. 32 p. für Technische Zusammenarbeit (GTZ)- Integrated Pest Management Löhr, B. 1997. The citrus wooly whitefly, (IPM) Horticulture Project, Nairobi, a new pest in eastern and southern KE. 32 p. Africa. Agrofor. Today Oct-Dec: 21-22.

140 Tanzania

CHAPTER 2.4 Tanzania

Ignas Swai* and Simon Slumpa**

Introduction Lake Victoria

Geographical context Kagera Mara Mwanza Arusha The following regions were covered in Moshi Shinyanga the surveys conducted during the Kilimanjaro 1997-99 period: Arusha, Babati, Kigoma Arumeru, Dodoma, Kongwa, Mpwapwa, Tanzania Tanga Njombe, Iringa, Mwanga, Rombo, Tabora Dodoma Zanzibar Singida Moshi, Hai, Same, Mkuu, Mbeya, Rukwa Iringsa Vwawa, Tukuyu, Morogoro, Tanga, Pwani Lushoto and Zanzibar (Figure 1). Mbeya Surveys followed the methodology Morogoro Njombe Lindi agreed among the project partners. Towns Mtwara Ruvuma The surveyed areas can be grouped Surveyed into distinct eco-climatic zones areas (Table 1). The northern highlands cover the Arusha and Kilimanjaro regions Figure 1. Regions covered in the 1997-99 survey, Tanzania. and Usambara Highlands in the Tanga region. The zone experiences two rainy seasons: short rains (vuli) from parts of Morogoro regions, where the November to December and long rains rainfall pattern is variable and (masika) from March to May. The main unpredictable. The Morogoro region is dry season is from June to September. transitional and receives higher rainfall The coastal lowlands include such than the Dodoma region. Dodoma areas as Muheza, Tanga, Ruvu, Dar-es- experiences only one rainy season, Salaam, Kibaha and Zanzibar Island. around November to January, and The rainfall pattern is the same as for production of vegetables in this zone is the northern highlands but the seasonal. Irrigation is common in both lowlands are hotter. The Central Morogoro and Dodoma. Plateau includes Dodoma and western The emergence of Bemisia tabaci as a pest and virus * Horticultural Research and Training vector Institute (HORTI-Tengeru), Arusha, Tanzania. The whitefly species Bemisia tabaci ** Selian Agricultural Research Institute (SARI), (Gennadius), the vector of Tomato leaf Arusha, Tanzania. curl virus (ToLCV) and Tomato yellow

141 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. Mean annual temperature, precipitation and altitude range of the surveyed regions in Tanzania.

Region (Province)a Mean min.-max. Mean Altitude temperature precipitation range (°C) (mm) (m)

Arusha 11-21 913 485-1660 Dodoma 16-29 556 329-378 Iringa 14-25 633 396-561 Kilimanjaro (Same) 14-26 1130 244-1600 Kilimanjaro (Moshi) 16-26 549 120-850 Mbeya 10-27 1132 424-536 Morogoro 19-30 854 450-900 Tanga (Lushoto) 19-30 1006 366-549 Zanzibar 21-30 1436 20-60 a. Same is on the windward side and Moshi, the leeward side, of Mt. Kilimanjaro.

leaf curl virus (TYLCV), is ranked as the characterized tomato geminiviruses of most important insect pest of tomato the Old World occur in Tanzania and (Lycopersicon esculentum Mill.) in was tentatively named Tomato leaf curl Tanzania (Varela and Pekke, 1995). B. virus-Tanzania (TLCV-Tan; Chiang et al., tabaci also transmits other viral 1997), now ToLCV-Tz. diseases in common bean (Phaseolus vulgaris [L.]), cowpea (Vigna unguiculata [L.] Walp.), sweetpotato (Ipomoea Increased Biological batatas [L.] Lam.) and cassava (Manihot Understanding esculenta Crantz), while the woolly whitefly Aleurothrixus floccosus (Maskell) is widespread in citrus (Citrus Characterization of spp. L.) growing areas. No information begomoviruses and whitefly was available on the geographical range biotypes of whitefly species and their socio- During the 1997-99 period, 179 samples economic significance in Tanzania prior of whitefly adults and nymphs were to the country’s participation in the collected from vegetable cropping Tropical Whitefly Integrated Pest systems in Tanzania, of which 407 Management (TWF-IPM) Project. nymph specimens were histologically processed and mounted on 213 slides. ToLCV was first reported in Identification of whiteflies conducted by Tanzania in 1990 (Czosnek et al., 1990). the International Center of Insect Additional survey studies conducted by Physiology and Ecology (ICIPE) revealed the Asian Vegetable Research and that B. tabaci was the most common Development Centre (AVRDC)-Africa whitefly species, representing 78% of the Regional Program in Tanzania from mounted specimens. B. afer (Priesner 1994-97 showed that several tomato and Hosney) represented 8% and samples with typical leaf curl symptoms Trialeurodes vaporariorum (Westwood), did not hybridize with the Egyptian or 7%. Other whitefly species made up 7% Israeli ToLCV-DNA probes. Polymerase of the specimens and included chain reaction (PCR) analysis performed Trialeurodes ricini (Misra), B. hirta Bink- with the above tomato samples Moenen, Orchamoplatus citri (Takahashi), indicated that in addition to TYLCV, a Tetraleurodes andropogon (Dozier) and virus different from all previously Aleurothrixus floccosus (Maskell).

142 Tanzania

The reproductive host plants of ranging from 1%-19%. The lowest whiteflies identified from the surveys in incidences of ToLCV symptoms, Tanzania belong to the following 1%-2%, were observed in the southern families: Amaranthaceae, Asteraceae, part of Tanzania (Iringa, Mbeya, Tukuyu Commelinaceae, Convolvulaceae, and Njombe) and in Zanzibar. There Cucurbitaceae, Euphorbiaceae, were no distinct ToLCV-free zones found Lamiaceae, Leguminosae, Myrtaceae, within the surveyed areas of Tanzania. Rutaceae, Solanaceae and Verbenaceae. ToLCV incidence was highest during the Table 2 gives the scientific names of the summer months (hot season), from host plant species on which each December to March. Yield loss due to whitefly species was found to be ToLCV depends on the susceptibility of reproducing. Whiteflies were abundant the tomato cultivar, the crop stage at on common bean but no whitefly- the time of infection and environmental transmitted disease symptoms were conditions (Nono-Womdim et al., 1999). observed. Twenty-three tomato accessions, ToLCV symptoms were observed in including progenies of crosses between Achyranthes aspera L., Euphorbia Lycopersicon chilense Dunal LA 1969 heterophylla L. and Nicandra and cultivated tomato, were field- physalodes (L.) Gaertn. Nono-Womdim screened for resistance to ToLCV by et al. (1996) identified TYLCV, the AVRDC (Nono-Womdin et al., 1999). causative virus in these hosts. These Two commercial F1 hybrids, Fiona and non-cultivated host plants are Tyking, had the highest level of widespread in the country and may resistance. Plants of these varieties were serve as major reservoirs of TYLCV. symptomless and they did not hybridize with Israel TYLCV-DNA probe. The Disease incidence and symptom varieties PSR-403511 and PSR-407111 severity were also resistant to ToLCV but hybridization tests showed that a few Areas identified as “hot spots” for symptomless plants contained ToLCV ToLCV incidence in Tanzania include DNA. The following progenies of crosses Morogoro, Dodoma, Kilimanjaro between L. chilense LA 1969 and (especially Same) and Arusha (Figure 1). L. esculentum had 30%-60% resistant Fifteen percent of the surveyed farms plants: Chilytlc 94-1, 94-2, 94-4, 94-5, had 100% ToLCV incidence in their 94-6 and MultichilTLC. Tests on two tomato crop. It should be highlighted commonly grown commercial tomato that all surveyed farms in Morogoro varieties, Moneymaker and Roma, (Dakawa, Kariakoo, Kipela, Kibundi, showed 100% susceptibility. Kiruka and Bigwa) had 100% TLC incidence. Up to 100% of the crop also were reported to be affected in Natural enemy species Kilimanjaro (Marwa, Bangalala and Samples of whitefly parasitoids and Same) and Dodoma (Mbalala, Chikula predators were collected during the and Mpwapwa). Nine percent of the surveys. Preliminary identification of the surveyed farms had ToLCV incidences sampled parasitoids showed that most ranging from 70% to 99%. These were were Encarsia sophia (Girault and in Arusha (Baraa), Kilimanjaro Dodd), while Eretmocerus sp. was (Mijongomeni, Kivulini and Mkolowoni) collected on only two occasions. A and Dodoma (Msolota). Ten percent of number of predators, coccinellids and the surveyed farms had ToLCV predatory bugs were also collected in the incidences ranging from 50% to 69%, vicinity of whitefly populations in the while 65% of the farms had incidences survey fields.

143 Whiteflies and Whitefly-borne Viruses in the Tropics L L. L. L. Cav. Cav. L. L. (L.) R. Br. L. L. sp. sp. sp. Euphorbia heterophylla Euphorbia heterophylla Achyranthes aspera Galinsoga parviflora Plant unidentified Plant unidentified Euphorbia heterophylla Tephrosia Leonotis nepetifolia Lantana camara Commelina Galinsoga parviflora Ageratum conyzoides Bidens pilosa Achyranthes aspera Amaranthus Euphorbiaceae Euphorbiaceae Lamiaceae Verbenaceae Mill. L. Leguminosae L. L. L. L. Crantz Asteraceae Crantz Crantz (L.) Walp (L.) Walp. (Thunb.) Matsum & Nakai Amaranthaceae L. Commelinaceae (L.) Lam. Amaranthaceae sp. sp. Euphorbiaceae sp. Asteraceae spp. Manihot esculenta Desmodium Phaseolus vulgaris Citrus Manihot esculenta Phaseolus vulgaris Desmodium Vigna unguiculata Phaseolus vulgaris Citrullus lanatus Lycopersicon esculentum Solanum tuberosum Psidium guajava Phaseolus vulgaris Manihot esculenta Ipomoea batatas Desmodium Vigna unguiculata Leguminosae Euphorbiaceae Euphorbiaceae Leguminosae Leguminosae Cucurbitaceae Solanaceae Myrtaceae Euphorbiaceae Leguminosae Convolvulaceae Family Species Family Species Hosney) (Westwood) (Dozier) (Maskell) Rutaceae (Takahashi) (Misra) (Gennadius) Bink-Moenen (Priesner and Bemisia hirta Orchamoplatus citri Tetraleurodes andropogon Aleurothrixus floccosus Trialeurodes vaporariorum Trialeurodes ricini Bemisia afer Bemisia tabaci Whitefly species Crops Non-cultivated hosts Table 2. Reproductive host plants of whitefly species encountered during surveys in Tanzania.

144 Tanzania

Geographical range of whitefly collected only from citrus. The few and whitefly-transmitted virus specimens identified as B. hirta require infestation additional confirmation. B. tabaci was the most common whitefly species throughout the Up to 100% ToLCV incidence was found at altitudes ranging from 260 to country (Table 3). B. afer was common in cassava in the areas of Iringa, 1060 m, while ToLCV incidence of up Arusha, Mbeya and Kilimanjaro to 70% was encountered as high as 1460 m altitude (Table 4). B. tabaci (Moshi). T. vaporariorum was encountered in Arusha, Dodoma, was the prevalent whitefly species up Iringa, Kilimanjaro (Same), Mbeya and to an altitude of 1260 m, while the two Trialeurodes species mentioned Tanga (Lushoto) but not in Morogoro, Kilimanjaro (Moshi) or Zanzibar. above were prevalent at higher T. ricini was common in Arusha, altitudes (Table 4). Iringa and Mbeya. A. floccosus was

Table 3. Species composition (%) of whitefly samples collected in regions of Tanzania.

Provincea Species compositionb

B.t. B.a. T.v. T.r. B.h. O.c. T.a. A.f.

Arusha 48 18 8 14 2 0 0 10 Dodoma 90 2.5 2.5 5 0 0 0 0 Iringa 37.5 25 25 12.5 0 0 0 0 Kilimanjaro (Same) 96 2 2 0 0 0 0 0 Kilimanjaro (Moshi) 89 11 0 0 0 0 0 0 Mbeya 40 17 33 10 0 0 0 0 Morogoro 99 0 0 0 0 1 0 0 Tanga (Lushoto) 91 2 5 2 0 0 0 0 Zanzibar 83 0 0 0 0 0 17 0 a. Same is on the windward side and Moshi, the leeward side, of Mt. Kilimanjaro. b. B.t., Bemisia tabaci; B.a., B. afer; T.v., Trialeurodes vaporariorum; T.r., T. ricini; B.h., B. hirta; O.c., Orchamoplatus citri; T.a., Tetraleurodes andropogon; A.f., Aleurothrixus floccosus.

Table 4. Maximum incidence (% of sampled plants showing disease symptoms) of tomato leaf curl (TLC) encountered and species composition (%) of whitefly samples collected at different altitudes in Tanzania.

Altitude (m) Maximum Whitefly species compositiona TLC incidence B.t. T.v. T.r. B.a. A.f. B.h. T.a. O.c. 0-60 4 83 0 0 0 0 0 17 0 60-260 100 91 9 0 0 0 0 0 0 260-460 100 88 3 4 5 0 0 0 0 460-660 100 66 18 7 9 0 0 0 0 660-860 100 84 0 0 14 0 2 0 0 860-1060 100 44 0 0 0 56 0 0 0 1060-1260 50 75 13 12 0 0 0 0 0 1260-1460 70 0 0 100 0 0 0 0 0 1460-1660 15 14 43 0 43 0 0 0 0 a. B.t., Bemisia tabaci; T.v., Trialeurodes vaporariorum; T.r., T. ricini; B.a., B. afer; A.f., Aleurothrixus floccosus; B.h., B. hirta; T.a., Tetraleurodes andropogon; O.c., Orchamoplatus citri.

145 Whiteflies and Whitefly-borne Viruses in the Tropics

Increased Socio-economic Producers ranked their pest and Understanding disease problems in tomato and associated vegetable crops in Tanzania (in order of importance) as follows: late Farmers’ assessment of blight (Phytophthora infestans [Mont.] whitefly-related problems de Bary), TLC, fruit borers (Helicoverpa Tomato is one of the most important armigera [Hübner]), whiteflies vegetables cultivated in Tanzania. Most (Homoptera: Aleyrodidae), aphids (79%) of the tomato producers (Aphididae), Fusarium wilt, early blight interviewed used their own land for (Alternaria solani [Ell. and Mart.] Jones production, while 17% used rented and Grout), red spider mite land. A vast majority of the producers (Tetranychus urticae Koch.), diamond (92%) were men. Almost all the back moth (Plutella xylostella [L.]), producers (96%) regarded tomato as angular leaf spot (Pseudomonas their most profitable vegetable crops syringae pv. Lachrymans), bruchids and 62% of producers have been (Coleoptera: Bruchidae), bacterial leaf involved in vegetable farming for more spot (Xanthomonas campestris pv. than 5 years. Although tomato visicatoria), leaf miner (Lyriomyza spp.), production has increased in the last beanfly (Ophiomyia spp.), cutworm 5 years to meet increased demand for (Spodoptera litura [Fabricius]), powdery processing and fresh markets, the mildew (Leveillula taurica), black rot average yields remain low at about (X. campestris pv. campestris), fruit fly 10-14 tons per hectare. (Drosophila melanogaster), bacterial canker (Clavibacter michiganensis), The main tomato varieties bean rust (Uromyces phaseoli), Tomato cultivated in Tanzania are Marglobe, mosaic virus, Septoria leaf spot (Septoria Moneymaker and Roma VF, all of lycopersici), nematodes (Meloidogyne which are highly susceptible to several spp.), Cucumber mosaic virus, thrips viral diseases. Most producers (60%) (Thysanoptera: Thripidae), cabbage obtain planting material from the local sawfly (Nematus spp.) and soft rot market, 36% use their own seeds and (Erwinia carotovora). 4% get planting material from other tomato producers. Other commonly Most producers (90%) were able to grown vegetable crops include cowpea, recognize the whitefly and 84% also pea (Pisum sativum L.), common bean, recognized ToLCV. Only 11% of the cabbage (Brassica oleracea L. var. interviewed producers knew that capitata L.), onion (Allium cepa L.), Irish whiteflies and ToLCV were interrelated. potato (Solanum tuberosum L.), None of the interviewed producers from eggplant (Solanum melongena L.), sweet Morogoro (where the disease incidence pepper (Capsicum annuum L.), okra was 100% on all surveyed farms) knew (Abelmoschus esculentus [L.] Moench), the cause of ToLCV; they did not have sweetpotato, Amaranthus spp., names for the whiteflies other than watermelon (Citrullus lanatus [Thunb.] “insects”, nor did they have precise Matsum. & Nakai), melon (Cucumis names for ToLCV. Most producers melo L.), cucumber (Cucumis sativus L. (75%) believed that the whiteflies var. sativus) and other cucurbits. caused problems on their farms. Whiteflies also attack most of these Almost half of them (43%) believed that crops. The great majority (94%) of the both whiteflies and ToLCV caused producers interviewed practice crop problems, whereas 30% attributed the rotation. problems to ToLCV only and 27% to whiteflies only.

146 Tanzania

Local names given to whiteflies producers mostly attributed the included chawa (lice), inzi weupe (white increase in severity to weather flies), kibanda, kifizi, kipe weupe, changes, particularly long dry spells. kipepeo (small butterflies), kurukury, Other factors mentioned by some mbuu, msubi weupe, mvumuu, ndaka, producers were ineffective insecticides, sughru (small flying insects), sunhuu, resistance to insecticides and old crop suru, tukorokotwa, wadudu (insects) remains (acting as a source of and wadudu weupe (white insects). The infestation and virus reservoir). local names given to ToLC included rasta (dreadlocks), ugonjwa wa kukunja Producers gave estimation of the (curling disease), ukoma (leprosy), costs involved in crop protection dume (sterile), ghojo, kibangi, kudulala, measures (see below). The costs were kutu (rust), bondia, majani, masai, mainly incurred in the purchasing of mdamango, mwanga bondia, ngofu, chemical insecticides and labor for ngumi (boxer), and kibangi and kobe their application. (tortoise). Costs estimated by producers Estimation of disease in control of whitefly/TYLCV incidence and yield losses complex per hectare of tomato Among the surveyed farms, 99% had US$ % producers TLC symptoms in their tomato crop 0-49 40 and 34% had ToLCV incidences above 50-99 18 25%. Common symptoms include leaf 100-199 26 curling, yellowing, chlorosis of leaf 200-299 9 margins, leaf distortion, reduction in 300-400 7 leaf size, shortening of the internodes, stunting, excessive branching and Pesticide use flower abscission (Nono-Womdim et al., One-third of producers (35%) received 1999). The average perceived yield loss recommendations on management in tomatoes due to the whitefly/ToLCV practices from technical advisors, 35% complex was 45%. Some producers from other tomato producers, 3% from (9%) reported a total yield loss, 13% sales agents, 10% from others and reported three-quarters yield loss, 32% 17% relied on their own judgement. reported half yield loss, 42% reported Among those producers who applied one-quarter yield loss and only 5% of insecticides, the decision on what the producers reported no yield loss. insecticides to apply and when was made by the producer in 93% of the Most of the producers interviewed cases, while 6% relied on technical (74%) believed that they had whitefly advice. and/or virus problems every year and 26% reported that the most severe Use of pesticides was the most problems occurred in 1997. Most common option for managing the producers (94%) also believed that whitefly/ToLCV problem and was there is a relationship between climate practiced by 76% of those interviewed. and whitefly/ToLCV occurrence. The Cultural control was practiced by 11% majority (90%) believe that the problem of the producers, while 13% did not is worst during the hot season, practice any control. The pesticides between December and March, while most commonly applied by vegetable the problem is less pronounced producers in Tanzania for control of between August and December. The whitefly/disease problems on their

147 Whiteflies and Whitefly-borne Viruses in the Tropics farms are profenofos, chlorpyrifos, and Dodoma were identified as the fenitrothion, dimethoate, pirimiphos- principal “hot spots” for ToLCV, where methyl, chlorpyrifos and diazinon incidences of symptoms of 100% were (organophosphates); deltamethrin, frequently encountered. Moderately lambda-cyhalothrin and fenvalerate high incidences were also observed in (pyrethoids); thiodan and endosulfan the Arumeru Highlands near Arusha (organochlorines); and methomyl and Kilimanjaro. (carbamate). Commercial cultivars of tomato Producers who made 10 insecticide grown in Tanzania are susceptible to applications per season to manage the TYLCV. Some breeding lines from the whitefly/ToLCV complex accounted for African Regional Program (ARP) of 21% of those interviewed. Those who AVRDC at Arusha have shown made from five to eight applications significant tolerance to TYLCV. AVRDC- comprised 31% of interviewees and a ARP in collaboration with the similar percentage made from one to Horticultural Research and Training four applications. However, 18% Institute (HORTI-Tengeru) have begun reported that they did not apply efforts to screen a broad array of insecticides. Most producers (62%) tomato germplasm and incorporate the applied insecticides as a preventive resistance genes into cultivated tomato measure, 26% made the application varieties. when they observed damage and 6% made the application according to the Further work in Tanzania, under calendar. the second phase of the TWF-IPM Project, should focus on evaluation of germplasm for use in breeding for Strengthened Research resistance to TYLCV and ToLCV, and Capacity development of IPM options for the management of the whitefly vector. This survey has substantially improved our understanding of the magnitude of the whitefly-associated problems in the References major vegetable production areas of Tanzania. The work conducted under Chiang, B. T.; Nakhla, M. K.; Maxwell, P.; this first, diagnostic, phase of the TWF- Schoenfelder, M.; Green, S. K. 1997. A new geminivirus associated IPM Project also has strengthened with a leaf curl disease of tomato in collaboration between the national Tanzania. Plant Dis. 81:111. agricultural research organization and international research institutions Czosnek, H.; Navot, N.; Laterrot, H. 1990. such as ICIPE and AVRDC. Geographical distribution of Tomato yellow leaf curl virus. A first survey using a specific DNA probe. Conclusions Phytopathol. Mediterr. 29:1-6. Nono-Womdim, R.; Swai, I. S.; Green, S. TYLCV is the most important whitefly- K.; Gebre-Selassie, K.; Laterrot, H.; transmitted viral disease in tomato Marchoux, G.; Opena, R.T. 1996. cropping systems in Tanzania. The Tomato viruses in Tanzania: disease causes significant yield loss in Identification, distribution and tomato production in various parts of disease incidence. J. S. Afr. Hort. the country. The lowlands of Morogoro Sci. 6:41-44.

148 Tanzania

Nono-Womdim, R.; Swai, I. S.; Green, S. K.; Chadha, M. L. 1999. Varela, A. M.; Pekke, A. 1995. Tomato yellow leaf curl virus and Proceedings of Tomato Planning Tomato leaf curl-like virus in Workshop for Eastern and Southern Eastern and Southern Africa. Africa Region, 16-20 October 1995, Paper presented at final IPM Harare, ZW. Deutsche Gesellschaft workshop held at the für Technische Zusammenarbeit International Center of Insect (GTZ)-Integrated Pest Management Physiology and Ecology (ICIPE), (IPM) Horticulture Project, Nairobi, Nairobi, KE. 2 p. KE. 32 p.

149 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 2.5 Malawi

Harriet Thindwa and Patric Khonje*

Introduction

Geographical context Karonga Malawi lies between latitude 9° 22' and 17° 7' South, and between longitudes Malawi 32° 40' and 35° 55' East. The climate is characterized by three main seasons: cool and dry from May to August, warm Mzuzu and dry from September to November, Lake and warm and wet from December to April. The 5-month rainy season (November to March) begins earlier in Kusungu southern and central regions of the country. Annual rainfall ranges from Kumuz Intl 600 mm in the lower Shire Valley and the Karonga Lake Shore Plains to over Lilongwe 3000 mm in high-elevation areas. Temperatures range between 20 °C and 35 °C but may approach and surpass Zomba 40 °C in the Rift Valley areas during Chileka Intl October and November (Mkanda et al., 1995). Dlantyre Towns

Surveys were carried out in the Surveyed main areas of the country growing areas tomato (Lycopersicon esculentum Mill.) during the period 1997 to 1999. Areas Figure 1. Tomato-growing areas surveyed for surveyed were Mzuzu, Mzimba, Nkhata whitefly incidence in Malawi. Bay, Salima, Lilongwe, Dedza, Ntcheu and Thyolo (Figure 1). The emergence of Bemisia tabaci as a pest and virus vector

In Malawi, the whitefly species Bemisia tabaci (Gennadius) has been reported as

* Bvumbwe Agricultural Research Station, the vector of cassava mosaic disease Limbe, Malawi. (Swanson and Harrison, 1994). Varying

150 Malawi levels of infestation by the citrus woolly Trialeurodes vaporariorum (Westwood), whitefly Aleurothrixus floccosus T. ricini (Misra) and Aleyrodes proletella (Maskell) also have been recorded in (Linnaeus) 1% each. The species regions of the country growing citrus composition of samples from the (Citrus spp. L.) (Löhr, 1996). various areas surveyed is reported in Table 1. Although whiteflies were suspected of transmitting viral diseases in Table 1. Species composition (%) of whitefly vegetable crops, especially tomato, no samples collected in tomato- reliable information on this subject was producing regions of Malawi. available from Malawi prior to the Province Species compositiona survey conducted under the Tropical B.a. B.t. T.v. T.r. A.p. Whitefly Integrated Pest Management Dedza 56 34 5 0 5 (TWF-IPM) Project. The occurrence of Lilongwe 86 14 0 0 0 whitefly species, their geographical Mzimba 86 14 0 0 0 distribution and their natural enemies Mzuzu 100 0 0 0 0 in the vegetable cropping system also were unknown. Nkhata Bay 53 47 0 0 0 Ntcheu 55 15 10 20 0 Salima 33 67 0 0 0 Increased Biological Thyolo 54 42 0 0 4 Understanding a. B.a., Bemisia afer; B.t., B. tabaci; T.v., Trialeurodes vaporariorum; T.r., T. ricini; A.p., Aleyrodes proletella. Characterization of begomoviruses and whitefly The reproductive host plants from biotypes which whitefly nymphs were collected belong to the families Solanaceae, Selection of survey sites and other Euphorbiaceae, Leguminosae and methodologies followed the standards Verbenaceae (Table 2). B. tabaci was agreed with other project partners. found breeding on tomato in only a few Questionnaires were used to gather areas. information from vegetable producers. The incidence of plants affected by Symptoms of tomato yellow leaf tomato yellow leaf curl was estimated curl were observed in most of the on their farms, whiteflies were sampled surveyed sites. Samples of about 10 for species identification and parasitoids plants per field were collected from all and suspected natural enemies were the affected farms, squashed on nylon collected. membranes and sent to the John Innes Centre (JIC) for hybridization. The During the surveys, 130 collections results confirmed the presence of of adults and nymphs of whiteflies were Tomato yellow leaf curl virus (TYLCV). made. Of these, 159 specimens were processed and mounted on 67 slides. Disease incidence and Preliminary identification carried out at symptom severity the International Center of Insect Physiology and Ecology (ICIPE) revealed Areas identified as “hot spots” (high that Bemisia afer (Priesner and Hosney) incidence of TYLCV) include Dedza, is the most prevalent species in the Mzimba, Thyolo and Ntcheu (Figure 1). country, accounting for 67% of samples The highest incidences of TYLCV were identified. Of the remaining specimens, observed at altitudes ranging from B. tabaci comprised 30%, and 1085 to 1485 m (Table 3).

151 Whiteflies and Whitefly-borne Viruses in the Tropics L. L. Euphorbia heterophylla Lantana camara Euphorbiaceae Merr. Verbenaceae Mill. (Bort.) Merr. (Bort.) L. L. L. L. Crantz Crantz Mucuna deeringiana Phaseolus vulgaris Manihot esculenta Mucuna deeringiana Phaseolus vulgaris Lycopersicon esculentum Phaseolus vulgaris Manihot esculenta Phaseolus vulgaris Leguminosae Leguminosae Leguminosae Euphorbiaceae Solanaceae Leguminosae FamilyEuphorbiaceae Leguminosae Species Family Species Hosney) (Westwood) (Linnaeus) (Misra) (Gennadius) (Priesner and Aleyrodes proletella Trialeurodes vaporariorum Trialeurodes ricini Bemisia tabaci Bemisia afer Whitefly species Crops Non-cultivated hosts Table 2. Reproductive host plants of whitefly species recorded in Malawi.

152 Malawi

Table 3. Maximum incidence (%) of Tomato yellow leaf curl virus (TYLCV) and species composition (%) of whitefly samples encountered at increasing altitudes in Malawi.

Altitude TYLCV Whitefly species compositionb a (m) incidence B.a. B.t. T.v. T.r. A.p.

485-685 14 31 56 0 0 13 685-885 14 38 63 0 0 0 885-1085 7 64 27 0 0 9 1085-1285 52 79 21 0 0 0 1285-1485 49 59 24 5 12 0 1485-1685 5 66 23 6 0 5 a. A TYLCV incidence of 94% was observed in one farm in the highest altitude range (1485-1684 m), while other farms in this altitude range had incidences of 0%-5%. b. B.a., Bemisia afer; B.t., B. tabaci; T.v., Trialeurodes vaporariorum; T.r., T. ricini; A.p., Aleyrodes proletella.

The survey revealed that B. afer is Increased Socio-economic the predominant whitefly species in Understanding most parts of Malawi (Table 1). This species was found on cassava (Manihot esculenta Crantz), common bean Farmers’ assessment of (Phaseolus vulgaris L.) and unidentified whitefly-related problems weeds but not on other vegetable crops. Most of the producers interviewed B. tabaci was also common in most (86%) grow tomato on their own land, parts of the country (Table 1). while 12% rent the land. The majority pT. vaporariorum was collected from of producers (88%) were men. Tomato Dedza and Ntcheu, while T. ricini was was regarded as the most profitable collected from Ntcheu only and vegetable crop by 98% of the A. proletella from Dedza and Thyolo. producers. Only 27% of them have High numbers of nymphs were been growing tomato for more than collected from Nkhata Bay, Dedza and 5 years. Tomato varieties commonly Ntcheu, the latter two being the only grown in Malawi include Moneymaker, regions from which more than three Red Khakhi, Vercles and Heinz. The whitefly species were identified. On tomato seed was bought from the local farms in Salima, very few whitefly market by 35% of the producers, 33% nymphs could be found for used their own seed (from the previous identification and the region had season), 21% obtained seed from the relatively low TYLCV incidences Agriculture Trading Company (ATC) (0%-11%). B. tabaci was the and 11% used imported seeds. Tomato predominant whitefly species up to an is grown both under rain-fed and altitude of about 900 m, while B. afer irrigated conditions. became more abundant at higher altitudes (Table 3). Other common vegetable crops in the country include common bean, Natural enemy species cabbage (Brassica oleracea var. capitata Nine collections of whitefly parasitoids L.), Irish potato (Solanum tuberosum and 10 of predators were made during L.), mustard (Sinapis alba L. subsp. the survey. Among these were Encarsia alba), onion (Allium cepa L.), pumpkin sophia (Girault and Dodd), coccinellids (Cucurbita spp.), rape (Brassica napus (Coleoptera: Coccinellidae) and L. var. napus), sweetpotato (Ipomoea predatory bugs (Heteroptera: batatas [L.] Lam.), turnip (Brassica Anthocoridae). rapa L. subsp. rapa), okra

153 Whiteflies and Whitefly-borne Viruses in the Tropics

(Abelmoschus esculentus [L.] Moench), Estimation of disease pepper (Capsicum annuum L.), eggplant incidence and yield losses (Solanum melongena L.), and carrot Among the surveyed farms, 75% had Daucus carota sativus ( L. subsp. TYLCV symptoms in their tomato crop. sativus [Hoffm.] Arcang. var. Hoffm.). However, only 6% of the farms had Cassava is also a major staple crop. incidences of TYLCV symptoms above Most producers (94%) practice crop 25%, while 16% of the farms had rotation. incidences ranging from 10% to 25%, and 53% of the producers had The main pests and diseases incidences ranging from 0.5% to 9%. recorded on tomato, ranked according The highest incidence (94%) was to importance, were late blight reported on a farm in Dedza. (Phytophthora infestans [Mont.] de Tetranychus Bary), red spider mites ( On average, the producers’ spp.) TYLCV, whiteflies, bacterial wilt estimate of yield loss in tomato due to Ralstonia solanacearum ( ) and cutworms the whitefly/TYLCV complex was 49%. Spodoptera litura ( [Fabricius]). According to the survey, 14% reported total yield loss, 14% reported three- Eighty-three percent of producers quarter yield loss, 35% reported half could recognize whiteflies and 82% yield loss, 29% reported one-quarter could recognize TYLCV. None of the yield loss and 9% did not report any producers seemed to know the losses due to the whitefly/TYLCV interrelationship between TYLCV and complex. Those who reported incurring whiteflies and confused TYLCV total loss were from the Dedza and symptoms with red spider mite Ntcheu Provinces, and a single damage, aphid damage, heavy rain producer from Mzimba. Producers who damage, fungal diseases, nitrogen reported abandoning tomato growing, deficiency and soil-borne diseases. at least once, due to the problem made Most of those interviewed (83%) up 21% of those interviewed; 38% of thought that whiteflies and TYLCV them abandoned their farms in 1997. caused problems on their farms. Almost half of the producers (46%) Most producers (68%) believed that complained about the whitefly only, they had whitefly/TYLCV problems 17% of the disease only and 37% every year. The most severe attack was complained about both the disease and in 1997 according to 43% of the the whitefly. respondents. These producers were from Ntcheu (Kasamba), Dedza Local names given to whiteflies (Kalilombe, Chimlambe, Jere), Nkhata include msambe, aphids, flies, pests, Bay (Kashonga, Chipayika), Salima vegetable lice, white aphids, cobweb, (Matumba), Mzimba (Chizuminja) and chinoni , white lice, white ants, mites, Mzuzu (Kaboko). Nearly all producers small flies, lice and white butterflies. (96%) believe there is a direct Local names given to TYLCV included relationship between the climate and chisaka blight, leprosy, (bushy top), whitefly/disease incidence. There was kaligwiti (bushy top), crinkled leaves, no obvious consensus among the katungana ciguduli , , bunchy top, producers as to what weather cicsaka kafumbata , and malformation. conditions trigger whitefly and TYLCV outbreaks. The whitefly/TYLCV problems were reported to be worst from January to June and again from September to October. Some producers

154 Malawi

(40%) believe that the whitefly/TYLCV people (3%) or a family member (2%), incidence is higher when it is dry and while 7% did not use insecticides on hot, while others (30%) said they their farms. Half the producers (52%) experience most problems when it is applied insecticides as a preventive wet and cold. The opinions of the measure, while 33% applied them remaining producers were scattered when damage was observed and 5% between the remaining combinations of applied insecticides according to the dry, moderate, wet and hot, moderate, calendar. Few producers (5%) reported and cold weather. making more than 10 insecticide applications per season, while 4% Producers gave estimation of the made 9-10 applications, 37% of the costs involved in crop protection producers made 5-8 applications, 40% measures (see below). made 1-4, and 14% applied no insecticides to manage the whitefly/ Costs estimated by producers in disease problem. control of whitefly/TYLCV complex per season of tomato production The insecticides most widely used US$ % producers in the management of the whitefly/ 0-49 26 TYLCV problem in Malawi include: 50-99 16 fenitrothion, malathion and dimethoate (organophosphates); cypermethrin and 100-199 19 200-299 14 lambda-cyhalothrin (pyrethroids); 300-400 9 carbaryl and sevin (carbamates). Farmers were ignorant of the fact that >400 16 chemicals with the same active Many producers (42%) had not ingredient often are sold under various trade names and they therefore risk received recommendations on management practices for whiteflies. using the same active ingredient Those who had received advice throughout, rather than alternating with others. The use of fungicides was obtained it from various sources: 34% received recommendations from common because producers confused technical advisors, 15% from other TYLCV symptoms with blight. Miticides such as dicofol were used also in an vegetable producers, 6% from family members or neighbours and 3% from attempt to control whiteflies but sales agents. Most producers (90%) EXTOXNET (1998) reports that this chemical has little effect on insects. managed the whitefly/virus problem by means of synthetic pesticides, 4% by cultural means and 6% practiced no control at all. Other control methods Strengthened Research such as legislative measures, natural Capacity insecticides (e.g., botanicals and bio- Scientists from ICIPE trained and pesticides), biological control and the accompanied the national scientists use of resistant varieties, were not during the beginning of the survey. The reported. project has equipped national scientists with a better understanding Pesticide use of the severity of whitefly-related The decision on what insecticide to problems in the country’s vegetable apply and when was made by the production. New information on the producer himself in 58% of the cases, status of whiteflies and the whitefly- by technical advisors (29%), sales transmitted viruses in tomato now is

155 Whiteflies and Whitefly-borne Viruses in the Tropics available. Understanding our encountered in all surveyed areas, producers’ perceptions and practices in indicating that the whitefly/TYLCV relation to whitefly problems has been complex is a potential threat to a significant achievement of this Malawi’s vegetable production. project. References Conclusions EXTOXNET (Extension Toxicology Whitefly/TYLCV “hot spots” in Malawi Network). 1998. Pesticide include Dedza, Mzimba, Thyolo and information profile, 3 Nov 1998. Available in: http:// Ntcheu. In both Dedza and Ntcheu, pmep.cce.cornell.edu/profiles/ several farmers reported having extoxnet/carbaryl-dicrotophos/ abandoned their tomato crops at least dicofol-ext.html (Accessed 2 May once due to whiteflies/TYLCV. For this 2000) reason, these two areas should be targeted for further research in the Löhr, B. 1996. The citrus woolly whitefly second phase of the project. B. afer, in eastern and southern Africa: Distribution, indigenous natural followed by B. tabaci, is the prevalent enemies and control strategies. In: whitefly species in most parts of the Procs. International Society of country and the most widespread Citriculture. Vol. 1, p. 619-624. parasitoid was E. sophia. Mkanda, F. X.; Kainja, S.; Kalindekafe, Although many producers M.; Chavula, G. M. 1995. Malawi: recognized whiteflies and TYLCV, they Greenhouse gas inventory and assessment of climate change often confused TYLCV symptoms with vulnerability. Interim report on those of other pest and disease or climate change country studies, nutrient deficiencies. Producers also March 1995. US Global Change were confused about the climatic Research Information Office conditions that trigger outbreaks of (GCRIO). Available in: http:// whiteflies and TYLCV. www.gcrio.org/CSP/IR/ IRmalawi.html (Accessed 11 May 2000) Incidences of TYLCV in Malawi were not as high as in neighbouring Swanson, M. M.; Harrison, B. D. 1994. Tanzania. Only 6% of the surveyed Properties, relationships and farms had TYLCV incidence above distribution of cassava mosaic 25%. Although TYLCV incidences were geminiviruses. Trop. Sci. 34:15-25. generally low, the disease was

156 Tomato Yellow Leaf Curl Virus in E. and S. Africa

CHAPTER 2.6 Tomato Yellow Leaf Curl Virus and Tomato Leaf Curl-like Viruses in Eastern and Southern Africa

Remi Nono-Womdim*, Ignas Swai*, Madan Mohan Lal Chadha** and Sylvia Green*

Introduction viruses account for most losses in farmers’ fields (unpublished results). Tomato (Lycopersicon esculentum Mill.) However, adequate information on the is one of the most popular vegetables in epidemiology of TYLCV and associated eastern and southern Africa. It is a viruses in eastern and southern Africa high-value crop, providing a good remain scanty. source of income to small-scale farmers. Tomato is consumed both in This chapter refers to the fresh and processed forms. The emergence of TYLCV and associated processing industry often provides good viruses in tomato in eastern and rural employment opportunities and southern Africa and describes the products for export. development of our understanding of their identification and management. The productivity of tomato in Africa, particularly in eastern and southern Africa, is unfortunately Impact and Assessment among the world’s lowest. For example, Losses the mean yield per hectare of tomato in southern African countries ranges from Over the past decade, TYLCV and leaf 1.5 to 14 tons per hectare as compared curl-like viruses have caused to the world average of 25 tons per significant losses for tomato farmers in hectare based on the 1989 Food and eastern and southern Africa. Yield Agriculture Organization (FAO) losses depend on the growing season, production estimates. Several fungal, tomato cultivars and growth stage at bacterial and viral diseases have which plants become infected. Recent drastically hampered the cultivation of on-farm surveys indicated that the tomato in eastern and southern Africa average perceived yield losses in tomato (AVRDC 1996; 1997; 1998). Among were up to 40% in Malawi, 50% in viral pathogens, Tomato yellow leaf curl Kenya, 75% in Tanzania and 100% in virus (TYLCV) and tomato leaf curl-like Sudan (CIAT, 1998). In Sudan, yield losses due to TYLCV have been reported to vary in different regions (Yassin and Nour, 1965; Makkouk et * Asian Vegetable Research and Development Center (AVRDC), Africa Regional Program. al., 1979). Moreover, Ioannau and ** Southern Africa Development Community Iordanou (1985) indicated that losses (SADC)-AVRDC-Collaborative Network for due to TYLCV vary between 50% and Vegetable Research and Development in Southern Africa (CONVERDS), Arusha, 82% depending on the time of Tanzania. infection.

157 Whiteflies and Whitefly-borne Viruses in the Tropics

In East Africa, TYLCV was first Diagnosis reported in the Sudan (Yassin and Nour, 1965). The disease has since In east and southern African countries, spread to Tanzania (Czosnek et al., TYLCV is primarily identified by the 1990; Nono-Womdim et al., 1996), presence of symptoms that consist of Malawi and Zambia (AVRDC, 1996). yellowing, leaf curling and stunting. Recent survey studies have indicated However, visual diagnosis is not useful that this virus occurs in Namibia and for the implementation of effective Swaziland (AVRDC, 1998), and in control strategies since the symptoms Kenya and Uganda (Table 1). A similar are very clear on old plants that are disease caused by a distinct virus, already fruiting. tentatively named “Tomato leaf curl virus-Tanzania” (TLCV-Tan; now Selected monoclonal antibodies accepted as ToLCV-Tz) (Chiang et al., developed against African cassava 1997) has been reported in the central mosaic virus (ACMV) or Indian cassava region of Tanzania (Nono-Womdim et mosaic virus (ICMV) have been used to al., 1996). detect different isolates of TYLCV (Macintosh et al., 1992; Givord et al., Table 1. Occurrence of Tomato yellow leaf curl 1994). These antibodies also were used virus (TYLCV) in eastern and to identify TYLCV isolates from Tanzania southern Africa. (unpublished results). It has been shown Country Year of surveya that many whitefly-transmitted 1996 1997 1998 begomoviruses are serologically related (Roberts et al., 1984). Therefore, Kenya 16/43 23/81 - serological tests for identification of Tanzania - 73/271 162/344 TYLCV are unspecific and do not Uganda - 5/35 10/58 differentiate strains or different leaf curl a. Number of infected samples/number of viruses. tested samples on which squash blot hybridization tests were performed using a specific DNA probe for TYLCV from Israel Nucleic acid hybridization assay is (TYLCV-Is). commonly used for the detection of TYLCV. Many cDNA probes prepared against different strains of TYLCV are Tomato plants infected with TYLCV now available (Czosnek et al., 1990; exhibit severe and varied symptoms. Kheyr-Pour et al., 1991). Some of these Common symptoms consist of leaf probes currently are being used to curling, yellowing, chlorosis of leaf identify leaf curl isolates from eastern margins, leaf distortion, reduction in and southern Africa. leaf size, shortening of the internodes or stunting. In some cases, infected plants show excessive branching and Etiology flower abscission. Symptoms of TYLCV vary with strain, tomato cultivars, TYLCV belongs to the Geminiviridae plant age at the time of infection and family and has small geminate particles environmental conditions. In eastern measuring about 20 × 30 nm (Harrison and southern Africa, epidemics of et al., 1977; Czosnek et al., 1988). The TYLCV occur during the summer genome of TYLCV consists of a circular months. single-stranded DNA encapsulated by a single capsid protein (Navot et al., 1991). Many isolates from eastern and southern Africa are closely related to TYLCV from Israel (TYLCV-Is). 158 Tomato Yellow Leaf Curl Virus in E. and S. Africa

TYLCV has been reported to have a less than 80% nucleotide identities wide host range, including plant with these other begomoviruses species belonging to Acanthaceae, (Chiang et al., 1997). Since it has been Asteraceae, Canicaceae, Euphorbiacea, proposed that two isolates of the same Leguminosae, Malvaceae, Oxalidaceae, begomovirus species have nucleotide Pedaliaceae, Plantaginaceae and sequence greater than 90%, the Solanaceae (Singh and Reddy, 1993). In Tanzanian begomovirus, TYLCV-Tz, is Tanzania, TYLCV was detected in three different from all previously newly identified hosts, Achyranthes characterized begomoviruses from the aspera, Euphorbia heterophylla and Old World. Recent survey studies have Nicandra physalloides (Nono-Womdim shown that TYLCV-Tz is now et al., 1996). These weeds are widespread in Tanzania (Table 2). A few widespread and serve as a major leaf curl samples from Malawi have reservoir of TYLCV. shown a weak positive reaction in squash blot hybridization with the DNA TYLCV is transmitted by the probe of TYLCV-Tz. These results whitefly Bemisia tabaci (Gennadius), not indicate that this virus, or another by seeds. In four African countries, closely related begomovirus, might be Kenya, Malawi, Tanzania and Sudan, present in this country. several populations of B. tabaci have been collected from TYLCV-infected Recently, another study conducted tomato samples and characterized on the identification of leaf curl viruses (CIAT, 1998). of tomato indicated the presence of a newly identified virus in Uganda that TYLCV diseases exhibit various has 73% homology with TYLCV-Is and symptoms indicating the existence of 78% homology with ToLCV-Tz. The leaf different strains. Preliminary curl virus from Uganda is tentatively identification studies conducted in named ToLCV-Ug (Charles Sskyewa, 1994 showed that several tomato personal communication, 1999). samples from Tanzania, with typical leaf curl and yellowing symptoms, did These findings indicate the not hybridize with the Egyptian and/or presence of at least three different leaf Israel DNA probes (Nono-Womdim et curl viruses of tomato in eastern and al., 1996). Two of these Tanzanian leaf curl samples were positive for begomovirus by polymerase chain Table 2. Preliminary results on the distribution of tomato leaf curl reaction (PCR) using a primer which disease in Tanzania using two specifically amplifies part of the specific DNA probes against tomato replicase (AC1) open reading frame yellow leaf curl virus-Egypt (TYLCV- Tomato leaf curl virus (ORF), the intergenic region and the Eg) and - Tanzania (ToLCV-Tz). coat protein (AV1) ORF of whitefly- a transmitted begomoviruses. Region Disease incidence Furthermore, the nucleotide sequence TYLCV-Is ToLCV-Tz of the PCR fragment was compared Arusha 13/40 39/40 with the sequences of several distinct Dodoma 79/113 96/113 begomoviruses, including TYLCV-Is, Kilimanjaro 12/30 29/30 TYLCV-Th, TYLCV-Sar, TYLCV-Ind, etc. Morogoro 34/70 60/70 that infect tomato, as well as with the Zanzibar 5/29 6/29 sequence of ACMV, ICMV and Mung a. Number of positive samples in DNA bean yellow mosaic virus (MYMV). It was hybridization tests/number of samples with found that the Tanzanian isolate shows leaf curl symptoms.

159 Whiteflies and Whitefly-borne Viruses in the Tropics southern Africa. The causes of this new Table 3. Field screening of Lycopersicon emergence of leaf curl viruses in Africa accessions for resistance to Tomato yellow leaf curl virus. have not been investigated yet. Lycopersicon Final disease accessions incidence (%)

Chepertylc C-1 96 Management 4-2 95 FI-3 100 Several management strategies to J-7 100 control TYLCV and associate viruses L-7 100 have been reported (Singh and Reddy, Sin-3 100 Siv-5 100 1993; Polston and Anderson, 1997). Siv-6 100 These strategies include cultural Columbian 36 100 practices, control of the insect vector EC-104395 100 and the use of resistant varieties. In Jackal 55 eastern and southern African LA 3214 100 LA 3216 95 countries, the implementation of Lignon C8-6 100 cultural practices such as use of virus- PSR-403511 5 free seedlings, reduction of virus PSR-407111 9 Roza 100 inoculum (roguing), elimination of Ty-20 54 sources of infection and intercropping 8476 75 to control TYLCV has not been Fiona 0 performed at farmers’ field level. Many Tyking 0 MoneyMaker 100 African farmers still cannot recognize Roma 100 TYLCV in order to apply these cultural practices for management. However, the use of insecticides, primarily to TYLCV and ToLCV-Tz since both control whiteflies and not TYLCV, is viruses occur in mixed infections in very common in eastern and southern Tanzania. Africa (CIAT, 1998).

Experiments have been conducted Conclusion since 1995 at AVRDC Africa Regional Program to identify tomato varieties This chapter summarizes recent with resistance to TYLCV (Table 3). The outbreaks of TYLCV and leaf curl-like varieties Fiona and Tyking are resistant viruses in eastern and southern Africa. to TYLCV in Tanzania. Plants of these The virus was first reported in Sudan species are symptomless and TYLCV is in the 1960s, where it has been a not detected by DNA hybridization. threat for several decades. Then, in Generally, resistance to TYLCV is quite 1990, it was found in Tanzania. To specific. For instance, tomato cultivars date, TYLCV is widespread in many with resistance or tolerance to TYLCV countries of the subregion. It is not from the Old World are susceptible to well known whether TYLCV has tomato leaf curl viruses from the occurred in these countries from some Western Hemisphere (Polston and time ago, as in Sudan, or whether it Anderson, 1997). The situation seems has spread southwards from to be different in Africa. In recent field Mediterranean countries. screening tests, Fiona, Tyking and TY52, which are resistant to TYLCV, Until 1997, TYLCV was the only were found to resist ToLCV-Tz as well. begomovirus reported to cause disease Such lines will provide an efficient on tomato in eastern and southern control strategy for the management of Africa. Two additional viruses, namely

160 Tomato Yellow Leaf Curl Virus in E. and S. Africa

ToLCV-Tz and ToLCV-Ug, which are Chiang, B. T.; Nakhla, M. K.; Maxwell, P.; different from TYLCV, have been Schoenfelder, M.; Green, S. K. reported recently (Chiang et al., 1997; 1997. A new geminivirus associated with a leaf curl disease of tomato in Charles Sskyewa, personal Tanzania. Plant Dis. 81:111. communication, 1999). Recent surveys indicate that many tomato leaf curl CIAT (Centro Internacional de Agricultura samples do not hybridize with the Tropical). 1998. Sustainable available cDNA probes in nucleic acid integrated management of whiteflies hybridization tests. This may indicate as pests and vectors of plant the existence of additional African viruses in the tropics. Progress report. Cali, CO. 131 p. indigenous leaf curl viruses. Czosnek, H.; Ber, R.; Antignus, Y.; Cohen, In Tanzania, tomato varieties with S.; Navot, N.; Zamir, D. 1988. resistant to TYLCV under field Isolation of Tomato yellow leaf curl conditions have been identified. These virus, a geminivirus. Phytopathology resistant tomatoes are hybrid varieties 78:508-512. that many farmers in eastern and southern Africa cannot afford. Czosnek, H.; Navot, N.; Laterrot, H. 1990. Geographical distribution of Tomato yellow leaf curl virus. A first survey At AVRDC, a breeding programme using a specific DNA probe. has been initiated aimed at Phytopathol. Mediterr. 29:1-6. incorporating the TYLCV resistance derived from Lycopersicon chilense Givord, L.; Fargette, D.; Kounounguissa, (Zakay et al., 1991) into cultivated B. R.; Thouvenel, J. C.; Walter, B.; tomato. However, until open-pollinate Van Regenmortel, M. H. V. 1994. Detection of geminiviruses from varieties with resistance to TYLCV tropical countries by a double become available, there is a need to monoclonal antibody ELISA using develop alternative control strategies antibodies to African cassava for African farmers that will include mosaic virus. Agron. 14:327-333. control of the insect vector and cultural practices. Harrison, B. D.; Barker, H.; Bock, K. R.; Guthrie, E. J.; Meredith, G.; Atkinson, M. 1977. Plant viruses with circular single-stranded DNA. References Nature 270:760-762.

AVRDC (Asian Vegetable Research and Ioannau, N.; Iordanou, N. 1985. Development Center). 1996. Epidemiology of Tomato yellow leaf Progress report 1995. Shanhua, curl virus in relation to the Tainan, TW. population density of its whitefly vector, Bemisia tabaci (Genn.). Tech. AVRDC (Asian Vegetable Research and Bull. 71, Agricultural Research Development Center). 1997. Institute, Nicosia, CY. 7 p. Progress report 1996. Shanhua, Tainan, TW. Khey-Pour, A.; Bendahmane, M.; Matzeit, V.; Accotto, G. P.; Crespi, S.; AVRDC (Asian Vegetable Research and Groenenborn, B. 1991. Tomato Development Center). 1998. yellow leaf curl virus from Sardinia Progress report 1997. Shanhua, is a whitefly-transmitted Tainan, TW. monopartite geminivirus. Nucleic Acids Res. 19:6763-6769.

161 Whiteflies and Whitefly-borne Viruses in the Tropics

Macintosh, S.; Robinson, D. J.; Harrison, Roberts, M. I.; Robinson, D. J.; Harrison, B. D. 1992. Detection of three B. D. 1984. Serological whitefly-transmitted geminiviruses relationships and genome occurring in Europe by tests with homologies among geminiviruses. J. heterologous monoclonal Gen. Virol. 65:1723-1730. antibodies. Ann. Appl. Biol. 121: 297-303. Singh, S. J.; Reddy, M. K. 1993. Leaf curl virus disease of tomato and its Makkouk, K. M.; Shehab, S.; Majdalani, management. Vatika from the Seed S. E. 1979. Tomato yellow leaf curl: and Plant People 2:5-21. Incidence, yield losses and transmission in Lebanon. Yassin, A. M.; Nour, M. A. 1965. Tomato Phytopath. Z. 96:263-267. leaf curl disease, its effect on yield and varietal susceptibility. Sudan Navot, N.; Pichersky, E.; Zeidan, M.; Agric. J. 1:3-7. Zamir, D.; Czosnek, H. 1991. Tomato yellow leaf curl virus: A Zakay, Y.; Navot, N.; Zeidan, M.; Kedar, whitefly-transmitted geminivirus N.; Rabinowitch, H. D.; Czosnek, H.; with a single genomic component. Zamir, D. 1991. Screening Virology 185:151-161. Lycopersicon accessions for resistance to the Tomato yellow leaf Nono-Womdim, R.; Swai, I. S.; Green, S. curl virus: Presence of viral DNA and K.; Gebre-Selassie, K.; Laterrot, H.; symptom development. Plant Dis. Marchoux, G.; Opena, R. T. 1996. 75:279-281. Tomato viruses in Tanzania: Identification, distribution and disease incidence. J. S. Afr. Hort. Sci. 6:41-44.

Polston, J. E.; Anderson, P. 1997. The emergence of whitefly-transmitted geminiviruses in tomato in the western hemisphere. Plant Dis. 81:1358-1369.

162 Conclusions and Recommendations

CHAPTER 2.7 Conclusions and Recommendations

Lisbeth Riis, Thomas Njuguna, Rebecca Raini, Berhard Löhr and Mohamed Ali Bob*

Results of Country (Priesner & Hosney), Siphoninus Studies phillyreae (Haliday), Aleurothrixus floccosus (Maskell), Trialeurodes ricini Bemisia tabaci (Gennadius) research in (Misra), Orchamoplatus citri the eastern African region has been (Takahashi), Tetraleurodes andropogon focused so far on cotton (Gossypium (Dozier) and Aleyrodes proletella hirsutum L.) in Sudan and on cassava (Linnaeus) were occasionally (Manihot esculenta Crantz) in Kenya, encountered in Malawi, Kenya and Tanzania and Malawi. In recent years, Tanzania, of which the last four species vegetables have become the crops most were registered for the first time in severely affected by B. tabaci- Tanzania. Bemisia hirta spec. nov. was transmitted viruses. Little attention encountered once but is subject to has been directed to vegetables confirmation. A. floccosus was therefore little information was encountered on citrus. In the East available prior to the commencement of African highlands, B. tabaci was the the System-wide Tropical Whitefly dominating whitefly species from sea Integrated Pest Management (TWF-IPM) level up to 1300-1700 m above sea Project in this region. For the first time, level, above which T. vaporariorum and information about critical aspects of T. ricini dominated. In southern Africa B. tabaci characterization on the (Malawi), B. tabaci only dominated up vegetable-based cropping system has to 900 m above sea level, above which been reported from the eastern Africa B. afer was the dominating species. region. Disease incidence and Characterization of symptom severity begomovirus and whitefly The Phase 1 survey revealed silver leaf biotypes symptoms on squash (Cucurbita pepo B. tabaci is the most dominating L.) in Sudan. Elsewhere, these whitefly species in the region as a symptoms are associated with B. tabaci whole. All specimens of whitefly biotype-B as the causal pest. This is collected in Sudan were B. tabaci; in the first observation of silver leaf Tanzania, B. tabaci made up 78% of symptoms in the region and may the collected specimens and in Kenya, indicate that this feared biotype-B has 65%. The species Trialeurodes reached Sudan. Characterization of the vaporariorum (Westwood), Bemisia afer B. tabaci specimens collected at the site where the silver leaf symptoms * International Center of Insect Physiology and were found is underway and the Ecology (ICIPE), Nairobi, Kenya. biotype-B is yet to be confirmed.

163 Whiteflies and Whitefly-borne Viruses in the Tropics

B. tabaci was found widespread on 11% in Kenya and 6% in Malawi. In tomato (Lycopersicon esculentum Mill.), the East African highlands, high ToLC pepper (Capsicum spp. L.) and eggplant incidence (>70%) was found from (Solanum melongena L.), in the 500-1000 m above sea level in Kenya Solanaceae; watermelon (Citrullus but from 50-1500 m above sea level in lanatus [Thunb.] Matsum. & Nakai), Tanzania. TYLC symptoms are melon (Cucumis melo L), squash, particularly severe during the hot cucumber (Cucumis sativus L. var. seasons (December to March in sativus) and bitter gourd (Momordica Tanzania and Kenya; March to charantia L.), in the Cucurbitaceae; September in Sudan). common bean (Phaseolus vulgaris L.), lablab (Lablab purpureus [L] Sweet), Reservoirs for TYLCV in the wild Leguminosae family; cassava vegetation were found in plant species (Euphorbiaceae); okra (Abelmoschus belonging to the families of Solanaceae esculentus [L.] Moench), cotton, and Euphorbiaceae, in particular, and sweetpotato (Ipomoea batatas [L.] Lam.), in Amaranthaceae. Desmodium and basil (Ocimum basilicum L.), in the Labiateae. In addition to In Sudan, the experienced team these families, B. tabaci also was found surveyed WTVs in all commonly grown frequently on wild plant species within vegetable crops. TYLCV in tomato is by the families of Acanthaceae, far the most economically important Amaranthaceae, Asteraceae, B. tabaci-transmitted virus in Sudan. Commelinaceae and Verberaceae. Next ranks Watermelon chlorotic stunt virus (WmCSV) and Okra leaf curl virus Whiteflies in association with (OLCV). TYLCV and WmCSV are Tomato yellow leaf curl virus (TYLCV) particularly severe during the hot had been studied in the Sudan (Dafalla season growing periods (March- and Sidig, 1997) and Tomato leaf curl September) and OLCV during the late virus (ToLCV) symptoms in Tanzania winter and early summer (January- had been found due to a unique strain March). Other severe WFTV were named ToLCV-Tz (Chiang et al., 1997). Tomato vein thickening virus (TVTV) in Since tomato was known as one of the tomato, Cucumber vein yellowing virus most profitable crops, and whiteflies (CVYV) and a yellowing-inducing poty- had been ranked as the most/second like virus not fully identified in most important pest/vector in tomato watermelon and musk melon (Cucumis (Varela and Pekke, 1995), tomato was melo L.). Less important WTVs, yet selected as the target crop in an representing a potential hazard, are extensive survey on whiteflies and TYLCV in pepper, WmCSV in snake whitefly-transmitted viruses (WTVs) in melon (Cucumis melo L. subsp. melo vegetable production in the region. var. flexuosus [L.] Naudin) and squash, a begomovirus and a closterovirus not Tomato yellow leaf curl (TYLC) yet fully identified in Vigna sp. and symptoms have been identified as common bean (Phaseolus vulgaris L. caused by TYLCV in Sudan, Kenya and var. vulgaris), Bean mild mosaic virus Malawi. TYLC/tomato leaf curl (ToLC) (BMMV) and Pepper leaf curl virus symptoms were found in more than (PepLCV). 93% of the surveyed fields of tomato in Sudan and Tanzania, in 72% in Malawi The African Regional Program (ARP) and 55% in Kenya. More than 25% of the Asian Vegetable Research and TYLC incidence was found in 50% of Development Center (AVRDC) has the fields in Sudan, 34% in Tanzania, identified three tomato accessions

164 Conclusions and Recommendations

(Fiona, Tyking and TY52) resistant to (Dedza, Mzimba, Thyolo and Ntcheu), TYLCV and ToLCV-Tz. Under heavy 600-1500 m. In Tanzania, Kenya and disease pressure, these accessions are Malawi the problem is most severe symptomless and the viruses are not during the hot dry season, December to detected by DNA hybridization. March. Irrigation in the semi-arid and Furthermore, two tomato accessions arid regions allows for continuous (PSR-403511 and PSR-407111) were cultivation of crops favoured by identified as tolerant to TYLCV; only B. tabaci and transmitted viruses, 5%-9% disease incidence was detected which guarantees continuous under heavy disease pressure. availability of reproductive hosts during the year. Natural enemy species Encarsia sophia (Girault and Dodd) was In hot spot areas throughout the the predominant whitefly parasitoid in region, yield losses due to B. tabaci and Kenya, Tanzania and Malawi, but was transmitted viruses ranged between not encountered during the survey in 50% and100%. As an average of all the Sudan. Encarsia lutea (Masi) and surveyed tomato fields, the yield losses Eretmocerus mundus Mercet. were due to B. tabaci and TYLC were 49% in predominant in the Sudan but not in Malawi, 45% in Tanzania and 30% in the other countries, although Kenya. In Sudan, the survey revealed Eretmocerus spp. were occasionally that an average of 62% fruit yields of encountered in Kenya and Tanzania. tomato, watermelon, musk melon and Encarsia formasa (Gahan) was okra are lost annually due to B. tabaci- encountered in the highlands of Kenya. transmitted viruses. WmCSV was Predators of coccinelids, lacewings, implicated in up to 80% yield losses of bugs, spiders and mites were found export Galia melon in Khartoum and adjacent to whitefly populations and Gezira States during the 1998-99 their identification at species level is season, and completely wiped out the yet to be confirmed. production in southern Blue Nile, Shuwak and Dinder inland delta. In Estimation of disease the region in general, more than 65% incidence and yield losses of producers reckoned that they had whitefly/disease problems in their In Sudan, the “hot spots” for B. tabaci tomato crop every year and 1997 was and transmitted viruses are the arid the year with the most severe attack and irrigated cotton-growing areas in (prior to the onset of the El Niño rains). Gezira and Khartoum States, at On average, producers in the region 400-500 m, where the problem is spend US$145 per hectare combating severe throughout the year but peaking whiteflies. in March to September. Where cotton is not grown, the incidence is less. In Tanzania, the hot spots for B. tabaci Farmers’ assessment of and TYLC are transitional areas in whitefly-related problems Morogor, Dodoma, Kilimanjaro and In Sudan, farmers ranked B. tabaci and Arusha, ranging from 100-1450 m. In its transmitted viruses as the most Kenya, the hot spots are semi-arid to economically and most damaging crop transitional areas in Eastern Province protection problem in tomato in the (Kibwezi, Kitui and Machacos) and Rift country. In Tanzania, farmers ranked Valley (Naivasha), 700-1200 m. In B. tabaci and its transmitted viruses as Malawi, the hot spots are in southern a second problem in tomato after areas of the Lake Malawi region bacterial blight. In Kenya and Malawi,

165 Whiteflies and Whitefly-borne Viruses in the Tropics

B. tabaci and its transmitted viruses Only 35% of the interviewed ranked less important. Nevertheless, producers in Sudan, 11% in Tanzania, 79% of the interviewed farmers in 1% in Kenya and none in Malawi knew Kenya found that the problems were on about the inter-relationship between the increase compared with 74% in whiteflies and ToLCV symptoms. In Tanzania, 68% in Malawi and 65% in Morogoro in Tanzania, where 100% Sudan. ToLCV incidence was found on all the surveyed farms, none of the producers It was generally believed that the knew about its inter-relationship with whitefly/disease complex is most whiteflies. In Malawi, almost all damaging during dry and hot seasons producers confused TYLC symptoms (December through March in Tanzania with other damage such as damage and Kenya, March through August in from red spider mite, aphids, heavy the Sudan). In Malawi, however, rain, fungal diseases, nitrogen controversial opinions on the deficiency and soil-borne diseases. seasonality of outbreaks were found among the interviewed producers. Sudan is the only country in the Interestingly, in the Sudan, TYLCV region that has adopted IPM as its incidence was most severe during the official policy of crop protection aimed at hot summer months despite less combating the whitefly problem. prevalence of the vector, whose Government support for IPM is well population peaked during the cooler defined in the recent publication, Sudan winter months (December through Country Strategy Note, 1997-2001: January). Partnership towards Sustainable Human Development. The document Pesticide use emphasizes that the Government of the Sudan is in pursuit of an integrated Eighty three percent of the farmers program for environmentally interviewed in the region applied sustainable development. However, pesticides in their tomato crops. The there is at present a laxity in percentage of farmers making more implementation and in cultural and than nine pesticide application per legislative measures. There are no tomato cropping season was 56% in restrictions on the number of pesticide Sudan, 21% in Tanzania, 9% in Malawi applications. Sudan has no extension and 8% in Kenya. The use of synthetic service at all, whereas technical insecticides was the most prevailing advisors from extension services have control method and in most cases the visited about one-third of the producer alone made the decision on interviewed producers in the other what insecticide to use and when countries. (average 84%). Half of the producers sprayed insecticides as a prevention measure and 25% of them would start spraying after damage had been Strengthened Research observed. Five to six sprayings per Capacity season was practiced on average but in The TWF-IPM Project has helped in the Sudan, 56% of the interviewed increasing the capacity in whitefly farmers made nine or more applications research by training one M.Sc. student per season. Pyrethroids and from Kenya and one Ph.D. student from organophosphates were the most Sudan. With assistance/input from the commonly used insecticides in the Institute of International Education region. (IIE), London, and the John Innes

166 Conclusions and Recommendations

Centre (JIC), project scientists were help understand whiteflies in future trained in conventional taxonomy and studies and training. DNA techniques. A short-term attachment also was completed on the This project demonstrates how effect of host plants on the oviposition effective collaboration can result in and survival of the sweetpotato whitefly. proper identification of research After long fruitless efforts, finally agendas and their correct cultures of Bemisia and Trialeurodes implementation. This has been possible were successfully established at the only because of the joint efforts of the International Center of Insect national programmes and the Physiology and Ecology (ICIPE). The collaborators in a well-coordinated Centre, AVRDC and the national approach. The joint efforts of the ICIPE partners are planning work on scientists and the national teams have transmission studies and host further strengthened ICIPE’s suitability studies in Phase 2. The hot partnership with national programmes spot areas of whitefly problems in in Africa. This mode of collaboration partner countries have been identified. and the complementation between Besides confirming the local occurrence ICIPE, AVRDC, JIC and national of several known natural enemies of agricultural research systems scientists whiteflies, additional genera/species of has helped link up the expertise of predators/parasitoids have been these centres and the local knowledge identified. Perceived on-farm losses due and experiences of the national to tomato geminiviruses and the programmes. The adoption of a average cost to control whitefly/TYLC standardized methodology and being per hectare of tomato in one season able to modify it to suit specific were documented. requirements has simplified work and made it possible to compare results and ICIPE has extended assistance in share experiences. the identification of whitefly species and natural enemy collections from surveys across nine countries in Africa to the Recommendations subproject on whiteflies on cassava and sweetpotato, led by the International As a result of a joint planning workshop Institute of Tropical Agriculture (IITA). in September 1999, the partners in Likewise, ICIPE has extended Sub-project 3 (Whiteflies as vectors of assistance in the identification of viruses in vegetable and mixed cropping whiteflies and natural enemies to systems in eastern Africa) made the countries that were not participating following recommendations for the officially in this project, for example, second project phase. Uganda and Senegal. The Center has established a reference collection for It was proposed that Phase 2 African whiteflies in the past 2 years should include Uganda in addition to from 10 countries within the framework Sudan, Tanzania and Kenya. The focus of the two African whitefly subprojects. should remain on B. tabaci as vector of This is in connection with a collection of viral diseases. The target cropping about 2000 samples of whiteflies and system in Kenya and Uganda will more than 4000 nymph specimens remain tomato. Target cropping systems mounted on permanent slides, in will include both tomato and addition to about 400 whitefly watermelon in Tanzania and tomato, parasitoids and predators collected. It is cucurbits and okra in the Sudan. There hoped that this collection will further is a need to continue the

167 Whiteflies and Whitefly-borne Viruses in the Tropics characterization of whiteflies and WTVs selection, multiplication and in target cropping systems. This dissemination. characterization comprehends both characterization at molecular level, Development of physical barriers including bio-typing of whiteflies, and for vector control was given second biological characterization of both priority in developing the IPM whiteflies and WTVs with emphasis on component. Different physical host range, host preferences and screening methods to prevent early interaction between the whiteflies and invasion of whiteflies should be the viruses. gathered from all over the world and tested as physical control measures The aim is to continue under the socio-economic and climatic strengthening the global whitefly conditions in each partner country. network set up by the TWF-IPM Project. Through this network, scientists from Testing rational use of conventional the region will enjoy the information and non-conventional insecticides was flow of knowledge and participation in given third priority in developing the experimental research in order to boost IPM component. Producers urgently the whitefly research in their respective need alternative control methods in countries with new knowledge and order to meet the new requirements of disseminate new control methods to the pesticide residue-free products for the national extension services and export market. Regional monitoring of vegetable producers. Initiatives taken to insecticide resistance of whiteflies will create an African Whitefly Network provide useful information to convince should continue. policy makers of the need for developing alternative control methods. More evidence must be recorded on Non-conventional and indigenous yield depression due to whiteflies and methods from all over the world such their transmitted viruses. Geographical as oil, non-phytotoxic soaps, botanicals information on agricultural regions, and ash should be gathered in a major crops, target crops, whitefly/ “package” and tested in partner virus hot spots, topographical and countries by means of a standardized climatic data, crop seasonality and methodology that will allow spatial patterns of cropping systems comparisons across ecosystems. must be collected in order to apply Evaluation of the effect of both geographic information systems to the conventional and non-conventional information created. control methods on natural enemies and biological control agents such as There is an urgent need for the parasitoids and entomopathogens will immediate development of IPM reveal information on their potential components viable under the socio- environmental impact and their economic conditions in the region. suitability in organic farming. Selection, multiplication and dissemination of tomato varieties Understanding of the mechanism resistant/tolerant to whitefly and/or behind effective cultural practices in ToLCVs were given first priority in the order to define appropriate cropping development of the IPM component. sequences also was put on the agenda. AVRDC should gather and multiply a Such studies may require several years “package” of resistant germplasm from and experimental cropping seasons all over the world and distribute it to and may not provide immediate the national partners for screening, applicable information for producers as

168 Conclusions and Recommendations the target stakeholders. The priority of Decision support tools must be such studies was therefore set to follow available for policy makers and news the above IPM components. Such media must be used currently as the studies would evaluate the effect of project develops. diversity-based cropping in time and space, planting dates, planting density, ground cover, roguing, in-field weed Acknowledgements management, crop residue management and micro-environment modification. The ICIPE-coordinated subproject has derived valuable input in geminivirus Studies on biological control of characterization from JIC in the UK and whitefly vectors were given less priority. from AVRDC in Tanzania. Dr. James Since low vector populations can create Legg (IITA, Uganda) offered valuable a severe virus outbreak, it is believed support in linking with the IITA-led that biological control may have low whitefly subproject on cassava/ impact in the control of virus outbreaks sweetpotato systems. The project also driven by the movement of whitefly has received valuable assistance from vectors. It was decided to start Dr. Andrew Polaszek (CAB International gathering cases of successful legal Bioscience, UK Centre) and Dr. Eddie control methods and related Ueckermann of the Plant Protection documentation. When sufficient Research Institute (PPRI), South Africa evidence has been provided through the in the identification of some whitefly project work, the aim is to start natural enemies. This book is a joint reinforcing official regulation and product of all partners and the ICIPE inspection for area-wide management, co-ordinating unit scientists. including planting date, harvesting date, crop free periods, destruction of crop residues, irrigation cut-off, etc. References

The research capacity in the Chiang, B. T.; Nakhla, M. K.; Maxwell, P.; Schoenfelder, M.; Green, S. K. 1997. partner countries must be strengthened A new geminivirus associated with a through training of national scientists leaf curl disease of tomato in in detection, identification and Tanzania. Plant Dis. 81: 111. characterization of whiteflies, WTVs, natural enemies of whiteflies and Dafalla, G. A.; Sidig, S. A. 1997. epidemiology. Furthermore, yearly Management of the whitefly-virus information exchange and co-ordination complex in Sudan. Food and Agriculture Organization (FAO) Plant meetings will be needed. Production and Protection Paper 143. Rome, IT. 214 p. When feasible, producers must be actively involved in the development of Varela A. M.; Pekke, A. 1995. Proceedings the IPM component, including the of the Tomato Planning Workshop brainstorming, planning, realization for Eastern and Southern Africa and evaluation of experiments. Training Region, 16-20 October 1995, Harare, ZW. Deutsche Gesellschaft für handouts on rational control of Technische Zusammenarbeit (GTZ)- whiteflies and WTVs must be produced Integrated Pest Management (IPM) for the extension service and producers, Horticulture Project, Nairobi, KE. and farmers’ organizations in hot spot 32 p. areas must be sensitized, in particular.

169 BLANCA 170 SECTION THREE

Whiteflies as Vectors of Viruses in Legume and Vegetable Mixed Cropping Systems in the Tropical Lowlands of Central America, Mexico and the Caribbean BLANCA 172 Introduction

CHAPTER 3.1 Introduction

Francisco Morales*

Establishing an created in 1992 under the initiative of International Network various international institutions such as the Centro Agronómico Tropical de Organizing an international network on Investigación y Enseñanza (CATIE) in whitefly-transmitted viruses in a region Costa Rica, the Pan American as extensive as Central America, Agricultural School at Zamorano in Mexico and the Caribbean is, for Honduras and the Instituto various reasons, no easy task. First, Interamericano de Cooperación para la whitefly-borne viruses have affected Agricultura (IICA). The network has more than a dozen countries and promoted annual meetings of regional commercial crops in this region. scientists interested in controlling this Second, most of these countries lack important pest, mainly under the the necessary research infrastructure coordination of CATIE. Another or resources to study plant viruses. important network that operates in the Third, most of the research conducted region is the Programa Cooperativo in the region has been focused on the Regional de Frijol para Centro América, whitefly Bemisia tabaci (Gennadius) México y El Caribe (PROFRIJOL), (Homoptera: Aleyrodidae) as a pest and financed by the Swiss Government to only rarely as a vector. And, finally, promote the production of common communication among researchers in bean, Phaseolus vulgaris L. the region remains difficult. However, (Leguminosae), in the region. A main this was the first task facing partners production problem of common bean in in Sub-project 2 of the Tropical the lowlands and the mid-altitude Whitefly Integrated Pest Management valleys of Central America, Mexico and (TWF-IPM) Project. Sub-project 2, the Caribbean has been Bean golden Sustainable integrated management of yellow mosaic virus (Geminiviridae: whiteflies as pests and vectors of plant Begomovirus), a disease caused by a viruses in the tropics, is led by the whitefly-transmitted virus. Both Centro Internacional de Agricultura PROFRIJOL and CIAT have given Tropical (CIAT) on behalf of the considerable attention to this problem Systemwide Programme on Integrated since 1978 through a collaborative Pest Management. research agreement. The existence of these networks greatly facilitated the Fortunately, a Regional Action Plan task of organizing an international for Whitefly Management had been network to focus specifically on viruses transmitted by the whitefly B. tabaci in * Centro Internacional de Agricultura Tropical legume and horticultural crops in the (CIAT), Cali, Colombia. region.

173 Whiteflies and Whitefly-borne Viruses in the Tropics

To further strengthen this network, four countries in Central America— Sub-project 2 selected three Guatemala, El Salvador, Honduras and subregional coordinators and three Costa Rica—were selected to conduct advanced research institutions with these studies, based on the following previous research experience in each of considerations. First, national the three subregions. The first institutions were available that were subregion selected was Central willing to conduct the surveys. Second, America and the advanced research a predominant small-scale farming institution chosen to provide technical system could be identified in the three support was the Plant Pathology countries and study areas selected. Department of the University of Third, a truly diversified agriculture, Wisconsin, Madison, USA. The including legumes and horticultural subregional coordinator was Dr. Luko crops, could be found in the study Hilje, an entomologist at CATIE, and areas selected. These conditions were the responsible scientist for the not satisfied in most of the areas University of Wisconsin was surveyed in Mexico or the Caribbean Dr. Douglas Maxwell. Subregion 2 was region, with the exception of Cuba, Mexico, and its support research which was excluded on the basis of institution was the University of other logistical constraints. Arizona, in Phoenix, USA. The subregional coordinator was Dr. Rafael Another consideration was that, in Rivera Bustamante of the Centro de Mexico and the Caribbean region, Investigaciones y Estudios Avanzados statutory measures to counter whitefly (CINVESTAV) in Irapuato Mexico, and and virus epidemics have been the responsible scientist for the successfully implemented. The national University of Arizona was Dr. Judith K. program scientists who have had the Brown. Subregion 3 was the responsibility of implementing the Caribbean, and its support institution regulatory measures will discuss the was the University of Florida, case of the Dominican Republic Bradenton, USA. The subregional (Chapter 3.11, this volume). coordinator was Dr. Colmar Serra, of the Instituto Superior de Agricultura (ISA), Dominican Republic, and the Emergence of Whitefly- responsible scientist at the University Transmitted Viruses of Florida, Dr. Jane Polston. Later on during the development of the project, A number of historical, economic and Drs. Hilje and Serra stepped down from biological factors have been identified their respective coordination roles and as being associated with the emergence the coordinator of Sub-project 2, of whitefly-borne viruses in legumes Dr. Francisco Morales, assumed these and horticultural crops in the region responsibilities. covered by Sub-project 2. These will be discussed in the individual country papers and in the concluding chapter Assessing Socio-economic of this section (Chapter 3.14, this and Environmental volume). However, in introducing this Impact section, two crosscutting trends can be highlighted. First, the presence of The large number of countries included whitefly-transmitted viruses is in Sub-project 2 precluded the frequently associated with the implementation of socio-economic cultivation of large-scale commercial or surveys in each country. Nevertheless, non-traditional horticultural crops,

174 Introduction especially for export. Second, the information system to analyse the data disintegration of national agricultural collected, and in the creation of an research programs throughout the interactive system to allow potential study region has had a negative impact users to access these data. The amount on the management of whitefly-related of data collected across this diverse production problems. Basically, the region considerably exceeded lack of technical assistance to control expectations and further analysis will the whitefly vector has left producers in be required to extract all the valuable the hands of pesticide vendors. As a information it contains. However, when consequence, pesticide abuse has this task is complete, the knowledge resulted, with serious biological and gained will provide a sound foundation environmental consequences. for developing and implementing sustainable management strategies for The economic situation in Latin whiteflies and whitefly-borne viruses in America has been deteriorating since the region. Figure 1 shows the areas the late 1970s, when the oil crisis affected by whitefly-transmitted generated a gigantic debt for most viruses. countries in the region. Possessing only an incipient industrial capacity, most In the following chapters, the countries had to resort to the export of reader will find a brief description of non-traditional crops such as soybean the different historical, geographic, (Glycine max [L.] Merr.) and vegetables. biological and socio-economic factors This strategy produced complex that contributed to the emergence of changes in the region’s agricultural the whitefly B. tabaci as a major pest environment that national agricultural and vector of viruses (referred to here research programs in Central America, as geminiviruses or, more specifically, Mexico or the Caribbean have not as begomoviruses). Concluding addressed. The cropping systems chapters draw together results involved have been difficult to analyse regarding the biotypes and in this project because the production reproductive hosts of B. tabaci in this losses caused by whitefly-transmitted region and, based on a synthesis of all viruses have forced growers to abandon results, make recommendations for and replace crops frequently. further research and implementation Nevertheless, the Sub-project partners directed towards achieving more have been able to document the rapid effective and sustainable management turnover of crops in whitefly-affected of whiteflies and whitefly-borne virus areas and to analyse some of the diseases in the region. consequences. Acknowledgements Looking Ahead The following chapters were written by A wealth of information, characterizing the senior author based on results problems associated with whitefly- obtained at the Plant Virology transmitted viruses affecting legumes Laboratory (CIAT, Cali, Colombia) from and horticultural crops in Central samples collected by the senior author America, Mexico and the Caribbean or collaborators in participating was accumulated during this countries, as well as on the available preliminary phase of the TWF-IPM literature and personal observations Project. Considerable progress was and experience in the region. Hence, made in the use of a geographic co-authors are not responsible for

175 Whiteflies and Whitefly-borne Viruses in the Tropics

Mexico Cuba

Haiti Honduras

Guatemala Nicaragua El Salvador Panama Costa Rica

Figure 1. Regions in the Neotropics affected by whitefly-transmitted viruses (shaded areas). omissions or errors found in these extensive survey: Claritza Gómez, Raúl chapters. Data provided by Sedano, Ana Karine Martínez, Ana collaborating national scientists are Cecilia Velasco, Mauricio Castaño, José dutifully acknowledged in the Arroyave, Iván Lozano, Natalia Villareal corresponding chapters. Several and Maritza Cuervo. Special thanks are national scientists and students due to Ms. María del Pilar Hernández, collaborating in the project conducted CIAT’s whitefly taxonomist, for socio-economic studies. Most of the examining all the whitefly specimens survey data presented have been collected during this investigation. Last published as theses by their main but not least, we recognize the work of authors, also included as co-authors in Dr. Lee Calvert, who supervised the the corresponding chapters. Special molecular characterization of B. tabaci acknowledgements are due to the biotypes and Trialeurodes vaporariorum professional personnel of the CIAT (Westwood) (Homoptera: Aleyrodidae) Plant Virology Laboratory, who specimens collected during this processed most of the samples of investigation. whiteflies and viruses analysed in this

176 Mexico

CHAPTER 3.2 Mexico

Francisco Morales*, Rafael Rivera-Bustamante**, Rafael Salinas Pérez***, Irineo Torres-Pachecoψ, Raúl Díaz Plazaψ, Wilson Avilés Baezaψ and Genovevo Ramírez Jaramilloψ

Introduction (Westwood) predominate. The latter species can become a major pest on some crops, including common bean Geographical context (Phaseolus vulgaris L.), tomato Mexico is the largest country in (Lycopersicon esculentum Mill.), potato Middle America, with an area of (Solanum tuberosum L.) and other 1,958,201 km2. The country can be horticultural crops grown above 900 m divided into different regions, the altitude; but it is not a vector of largest being the Central Plateau, begomoviruses. Figure 1 shows the which extends from the United States main agricultural regions affected by border in the north to the Isthmus of whitefly-transmitted begomoviruses. Tehuantepec in the south. At its northern end, the plateau is about The Central Plateau is flanked by 1320 m above sea level, rising to more the Sierra Madre, two mountain ranges than 2650 m south of Mexico City, the running west and east of the plateau. country’s capital. This Central Plateau To the south is the Balsas Depression, is divided into two parts—the Mesa del a hot, dry and broken area with Norte extends from the United States numerous small basins. Further south border to near San Luis Potosí, and the is the Mesa del Sur, with small isolated Mesa Central from San Luis Potosí to valleys, 1300 m to 1650 m above sea just south of Mexico City. The Mesa level. The Valley of Oaxaca is the Central is a less arid, higher and flatter largest and most densely populated of plateau than the Mesa del Norte. these valleys. The Southern Highlands Although the Central Plateau contains consist of a series of mountain ranges many agricultural basins and valleys, and plateaux, including the Sierra the altitude limits the survival and Madre del Sur. The Isthmus of activity of the whitefly Bemisia tabaci Tehuantepec is a narrow stretch of (Gennadius). Instead, other whiteflies lowland (up to about 330 m altitude) such as Trialeurodes vaporariorum with a hilly central area. The presence of B. tabaci south of the Central Plateau has not been well documented and it seems that T. vaporariorum may * Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. predominate as a pest of agricultural ** Centro de Investigaciones y Estudios significance. The Chiapas Highlands Avanzados, Irapuato, Mexico. are an extension of the mountain *** Instituto Nacional de Investigaciones ranges of Central America. The Sierra Forestales y Agropecuarias (INIFAP), Sinaloa, Mexico. de Soconusco runs along the Pacific ψ INIFAP, based at Mérida, Yucatán, Mexico. Coast. The Grijalva river valley leads

177 Whiteflies and Whitefly-borne Viruses in the Tropics

Mexico Sonora Baja California

Sinaloa Tamaulipas South Baja California San Luis Potosí Nayarit Yucatán

Veracruz Quintana Roo

Chiapas

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Mexico. northwest into the Tabasco Plain, an Tepic, Nayarit. Parts of these arid extension of the Yucatán Peninsula, regions have been irrigated since the bordering the Gulf of Mexico. This 1930s and support intensive cropping peninsula consists of flat, limestone systems of traditional and non- terrain, seldom exceeding 160 m above traditional (export) crops. Finally, Baja sea level. The northernmost coastal California is an isolated strip of arid area of the Yucatán Peninsula is an land, about 1280 km long and 160 km important horticultural area and has wide. Most of the eastern side of this been affected by B. tabaci and different peninsula is mountainous, reaching begomoviruses transmitted by this elevations of almost 3000 m and vector. sloping downwards towards the west.

The main horticultural areas of Mexico affected by B. tabaci are the The emergence of Bemisia coastal lowlands that lie east and west tabaci as a pest and virus of the Central Plateau. The Gulf vector Coastal Plain extends some 1400 km from the Texas border to the Yucatán The first report of B. tabaci as a vector Peninsula. The triangular northern of plant viruses in Mexico, in the early portion is more than 160 km wide near 1950s, is linked to the production of the US border, narrowing southwards cotton (Gossypium hirsutum L.) in the until it encounters the Sierra Madre Valley of Mexicali, Baja California Oriental close to the sea, north of (Cárdenas et al., 1996). The disease Tampico. This is a swampy area with observed was probably “cotton leaf several lagoons and is called the Zona crumple”, caused by a begomovirus Lagunera. South of this constriction, transmitted by B. tabaci, previously the Gulf Coastal Plain runs, narrow observed in the United States near the and irregular, all the way to the Colorado River Valley (Brown, 1994). Yucatán Peninsula. The narrower Around that time, B. tabaci was also Pacific coastal lowlands begin near the associated with a disease of tomato Mexicali Valley in the north and end, (Lycopersicon esculentum Mill.), called also some 1400 km to the south, near ‘chino del tomate’, in the Valley of

178 Mexico

Culiacán, Sinaloa, in north-western cultivation of melon (Cucumis melo L.), Mexico (Gallegos, 1978). In 1962, soybean (Glycine max [L.] Merr.), tomato B. tabaci was observed attacking cotton and peppers (Capsicum spp. L.), has in the Soconusco region of the southern created conditions favourable for the state of Chiapas (Cardenas et al., 1996). reproduction of B. tabaci away from its traditional hosts such as cotton. These In 1978, B. tabaci caused non-traditional export crops are considerable damage to horticultural currently planted in B. tabaci-affected crops in the Huasteca region, including regions such as the north-west, Las parts of the states of Hidalgo, Huastecas and Yucatán. Among the Tamaulipas, San Luis Potosí and most important viruses affecting Veracruz, where the Gulf Coastal Plain horticultural crops in Mexico may be tapers to form a narrow strip that mentioned: Tomato chino (leaf crumple) reaches all the way to the northernmost virus, described in Sinaloa in 1970 point of the Yucatán Peninsula. A (Gallegos, 1978); Serrano golden mosaic disease believed to be Bean golden virus, in north-eastern Mexico (Sánchez yellow mosaic virus (BGYMV) was et al., 1996) and Texas pepper virus observed for the first time in the north- (Stenger et al., 1990), both of which are western state of Sinaloa, in 1974 regarded as Pepper golden mosaic virus (López, 1974). When this disease (PepGMV); (Brown et al., 1993); Pepper occurred later (1990) in the state of Huasteco yellow vein virus (PHYVV) in Sonora, it was shown to be caused not pepper and tomato in Las Huastecas by BGYMV but by a distinct virus, Bean and Sinaloa (Hou et al., 1996); and calico mosaic virus (BCaMV) (Brown et “El Tigre” virus of pepper in Tamaulipas al., 1990). BCaMV induces more (Brown, 1989). In Yucatán, chilli pepper striking golden mosaic symptoms, (Capsicum spp.), tomato, squash which eventually resemble a bleaching (Cucurbita spp. L.), melon, and effect. The virus is not closely related to watermelon (Citrullus lanatus [Thunb.] BGYMV but rather to Squash leaf curl Matsum. & Nakai) have been attacked virus (SLCV), a virus previously since the 1980s by B. tabaci and the described attacking cucurbits in begomoviruses this whitefly species California (Flock and Mayhew, 1981). transmits (Díaz-Plaza et al., 1996). BGYMV emerged on the Gulf Coast of Veracruz in 1977. In 1979, it affected More recently, Tomato mottle virus common bean plantings in Las (ToMoV), a begomovirus originally Huastecas and in 1980 affected the described from Florida, United States, same crop in the state of Chiapas was detected in the Yucatán Peninsula, (López-Salinas, 1994). Finally, in 1980, also affecting tomato (Garrido and B. tabaci reached the Yucatán Gilbertson, 1998). In addition, PHYVV Peninsula, particularly the state of was found in Habanero varieties of chilli Yucatán, where it caused severe yield pepper and in pepper, and PepGMV was losses to horticultural crops as a pest found in Habanero. PHYVV had already and virus vector (Díaz-Plaza et al., been detected infecting Jalapeño 1996). varieties of chilli pepper in the neighbouring state of Quintana Roo Most of the subsequent (Díaz-Plaza et al., 1996). Meanwhile, an dissemination of whitefly-transmitted Old World begomovirus, Tomato yellow begomoviruses in Mexico has been leaf curl virus (TYLCV) has made its associated with the boom in appearance in the Yucatán Peninsula horticultural and other export crops (Ascencio-Ibáñez et al., 1999). This virus that took place in the 1980s. The has caused yield losses in the

179 Whiteflies and Whitefly-borne Viruses in the Tropics

Caribbean estimated at millions of conducts research on begomoviruses US dollars and thus poses an evident affecting horticultural crops in Mexico. threat to agriculture in Mexico. Dr. Rafael Rivera-Bustamante conducted surveys in collaboration with Dr. Irineo Torres-Pacheco, co-ordinator of the Advances in Biological horticulture program of the Instituto Research Nacional de Investigaciones Forestales y Agropecuarias (INIFAP) at that time. The participation of Mexican researchers in Sub-project 2 of the In the course of this project, samples Tropical Whitefly Integrated Pest were taken from crop plants showing Management (TWF-IPM) Project begomovirus-like symptoms at a range provided an opportunity to gain an of sites in major horticultural production updated and more systematic view of areas of Mexico (Table 1) and assayed the status of whitefly-borne viruses in using monoclonal antibodies (Cancino the country. The Centro de et al., 1995). In addition, a wide-ranging Investigaciones y Estudios Avanzados survey was conducted of viruses affecting (CINVESTAV), located at Irapuato, tomato and chilli in Mexico (Table 2). Guanajuato, was selected as the lead The results obtained demonstrate that institution in Mexico. This Centre begomoviruses can attack the main

Table 1. Analyses of plants affected by begomovirus-like diseases in Mexico.

Location Plant (year) Begomovirusa Other viruses MAB-BS MAB-GA

Los Mochis, Sinaloa Common bean (93) + - - Los Mochis, Sinaloa Common bean (93) + - - J. Rosas, Guanajuato Tomato + - - Dzemul, Yucatán Squash + - - Dzemul, Yucatán Chilli - - - Dzemul, Yucatán Chilli - - - Dzemul, Yucatán Chilli + - - Los Mochis, Sinaloa Common bean + - - Los Mochis, Sinaloa Common bean + - - Los Mochis, Sinaloa Common bean + - - Los Mochis, Sinaloa Squash + na - Los Mochis, Sinaloa Squash + na - Los Mochis, Sinaloa Squash + na - Culiacán, Sinaloa Common bean + na - Culiacán, Sinaloa Common bean + na - Culiacán, Sinaloa Common bean + - - Culiacán, Sinaloa Common bean + - - Culiacán, Sinaloa Soybean + na - Culiacán, Sinaloa Soybean + na - Culiacán, Sinaloa Squash + na - Culiacán, Sinaloa Squash - na - Etchojoa, Son. Common bean + - - a. MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses; MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of Bean golden yellow mosaic virus-Guatemala; and na, not analysed.

180 Mexico

Table 2. Virus survey of tomato and chilli fields in 22 states of Mexico.

Sample Plant Localitya State N W V1b V2b V3

1 Sida Chetumal Q. Roo 18 30 88 15 TMV TEV - 2 Chilli Chetumal Q. Roo 18 30 88 15 PHYVV TPV - 3 Malvaceae Campeche Campeche 19 50 90 26 TMV TEV - 4 Weed Uxmal Yucatán 20 25 89 42 TMV - - 5 Chilli Campeche Campeche 19 50 90 26 PHYVV - - 6 Chilli Sayula Jalisco 19 50 103 30 TPV - - 7 Chilli Sayula Jalisco 19 50 103 30 TEV - - 8 Tomato Sayula Jalisco 19 50 103 30 TMV PHYVV - 9 Squash Sayula Jalisco 19 50 103 30 PHYVV - - 10 Euphorbia Sayula Jalisco 19 50 103 30 TPV - - 11 Sida Sayula Jalisco 19 50 103 30 PHYVV - - 12 Chilli Tecomán Colima 19 01 103 55 TMV PHYVV - 13 Chilli Tecomán Colima 19 01 103 55 TMV - - 14 Chilli Tecomán Colima 19 01 103 55 PHYVV - - 15 Chilli Tecomán Colima 19 01 103 55 TPV - - 16 Weed Tecomán Colima 19 01 103 55 TPV - - 17 Chilli S. Ixcuintla Nayarit 21 50 105 10 PHYVV - - 18 Chilli S. Ixcuintla Nayarit 21 50 105 10 TPV - - 19 Tomato S. Ixcuintla Nayarit 21 50 105 10 TPV - - 20 Chilli Apaseo Alto Guanajuato 20 35 100 25 PHYVV - - 21 Weed Celaya Guanajuato 20 32 100 49 TMV - - 22 Malvaceae Los Angeles R. Lagunera 25 32 103 30 TMV - - 23 Common bean Los Angeles R. Lagunera 25 32 103 30 TMV - - 24 Tomate CIAN R. Lagunera 25 32 103 30 TMV - - 25 Melon CELALA R. Lagunera 25 32 103 30 TMV - - 26 Tomate CELALA R. Lagunera 25 32 103 30 TMV - - 27 Chilli Ciudad Isla Veracruz 18 5.1 95 36 TPV - - 28 Chilli Martinez dlt Veracruz 20 5.0 97 5.0 PHYVV TPV - 29 Chilli Papantla Veracruz 20 30 97 12 PHYVV - - 30 Chilli Papantla Veracruz 20 30 97 12 PHYVV TPV - 31 Chilli Poza Rica Veracruz 20 33 97 26 PHYVV - - 32 Melampodium Veracruz Veracruz 19 9.5 96 6.4 TMV CMV TEV/TPV 33 Chilli Guegovela Oaxaca 16 46 96 49 TEV - - 34 Datura S.L.Cacaote Oaxaca 17 7.9 96 47 TPV - - 35 Tomato Guegovela Oaxaca 16 46 96 49 TEV - - 36 Chilli Pabellón Aguascal. 21 55 102 15 CMV - - 37 Chilli Jalpa Aguascal. 21 55 102 15 TEV - - 38 Malvaceae Jalpa Aguascal. 21 55 102 15 TEV - - 39 Chilli Jaral Progr. Aguascal. 20 30 101 6 TEV TMV CMV 40 Chilli Jaral Progr. Aguascal. 20 30 101 6 TEV TMV - 41 Chilli Jaral Progr. Aguascal. 20 30 101 6 TEV CMV - 42 Chilli Paracuaro Guanajuato 20 15 100 48 TEV CMV - 43 Chilli Paracuaro Guanajuato 20 15 100 48 TEV CMV - 44 Malvaceae S. Luis Paz Guanajuato 21 33 100 32 TMV - - 45 Malvaceae S. Luis Paz Guanajuato 21 33 100 32 TEV - - 46 Tomato Yautepec Morelos 18 50 99 11 PHYVV - - 47 Physalis Yautepec Morelos 18 50 99 11 PHYVV - - 48 Chilli Yautepec Morelos 18 50 99 11 PHYVV TPV - 49 Tomato Metztitlán Hidalgo 20 38 98 39 PHYVV TPV - 50 Chilli Metztitlán Hidalgo 20 38 98 39 PHYVV - - 51 Chilli Salitral S.L. Potosí 22 53 102 02 PHYVV TPV - 52 Tomato Salitral S.L. Potosí 22 53 102 02 PHYVV - - 53 Sida Salitral S.L. Potosí 22 53 102 02 PHYVV - - 54 Chilli Apaseo Querétaro 20 35 100 25 PHYVV - - a. CIAN and CELALA are experiment stations in the Lagunera region of Mexico. b. TMV, Tobacco mosaic virus; TEV, Tobacco etch virus; PHYVV, Pepper Huasteco yellow vein virus; TPV, Texas pepper virus (= Pepper golden mosaic virus); CMV, Cucumber mosaic virus.

181 Whiteflies and Whitefly-borne Viruses in the Tropics horticultural crops grown in the state—and took samples of common country: common bean, tomato, chilli bean plants affected by symptoms of the and squash. Additionally, a begomovirus disease. A partial molecular was detected in soybean plants in characterization of the viral isolates Sinaloa State. Soybean is a host for collected (Table 3) shows that BCaMV is B. tabaci in South America but its role still present in common bean plantings in Mesoamerica as a reproductive host in north-western Mexico, specifically in for this whitefly seems to have been Los Mochis. This isolate is still closely secondary to that of other crops such related to an isolate of BCaMV collected as cotton and some horticultural crops. in the same locality in 1986. The sample Nevertheless, B. tabaci reportedly from Culiacán represents a begomovirus destroyed over 5000 ha of soybean in closely related to BCaMV but already the Valle del Fuerte, Sinaloa, in 1994 evolving into a new species (as indicated (INIFAP, 1995). by a sequence homology of less than 90%). Perhaps of even greater interest is As noted above, golden mosaic that the common bean begomovirus symptoms in common bean in northern from Sonora is still identifiable as an Mexico are caused by BCaMV, which is isolate of the original SLCV, constituting related to SLCV, a cucurbit virus the first direct evidence of the evolution originally recorded from the United of this virus. States. In this project, we surveyed three key locations in the north-western Selected whitefly samples were also region of Mexico—Etchojoa in Sonora, collected on different crops in north- and Los Mochis and Culiacán in Sinaloa western Mexico. Table 4 presents the

Table 3. Comparative sequence homologies (%) of common bean begomoviruses from north-western Mexico (Sonora and Sinaloa) and other whitefly-transmitted begomoviruses.

Virus isolate ORFa Common bean begomovirusesb

BGYMV-GA BCaMV SLCV BCaMV-MO BDMV

Etchojoa, Sonora AC1 57.0 77.2 96.0 76.8 61.0 AV1 72.2 75.4 97.8 74.9 74.2 Los Mochis, Sinaloa AC1 69.8 89.7 70.5 88.3 68.8 AV1 74.4 92.2 76.1 92.2 73.9 Culiacán, Sinaloa AC1 74.3 83.0 69.1 81.8 74.6 AV1 73.0 86.1 79.2 85.7 74.2 a ORF, open reading frame; AC1, replicase; and AV1, capsid protein. b BGYMV-GA, Bean golden yellow mosaic virus-Guatemala; BCaMV, Bean calico mosaic virus; SLCV, Squash leaf curl virus; MO, Mochis; BDMV, Bean dwarf mosaic virus.

Table 4. Identification of Bemisia tabaci biotypes from Sinaloa, northern Mexico.

Locality Crop Biotype A Biotype B

Los Mochis Common bean X - Los Mochis Common bean X - Los Mochis Common bean - X Los Mochis Common bean - X Los Mochis Common bean - X Culiacán Eggplant X - Culiacán Eggplant X - Culiacán Eggplant X - Culiacán Eggplant X - Culiacán Eggplant X -

182 Mexico results of the molecular biotyping of (2) a climate database and (3) a soils the whiteflies collected. As can be database. For the biological observed, there is a mixed population parameters, a mean monthly of biotypes A and B of B. tabaci. temperature of 23-27 °C and annual rainfall of less than 50 mm were Finally, the project supported work chosen as defining areas with a high conducted by INIFAP researcher, probability of B. tabaci infestation. Genovevo Ramírez-Jaramillo, based in Whitefly population dynamics in Mérida, Yucatán, on “the bio-ecology of Yucatán were influenced to a large the whitefly B. tabaci in the Yucatán extent by the amount of rainfall but Peninsula.” This work—which also only minimally by the temperature, included the Government of the State which remains within a range that is of Yucatán, the Secretary of favourable for whitefly development Agriculture, Livestock and Rural (Figure 1). Thus, the peak populations Development, the Regional Research of B. tabaci occurred in the months of Centre for the South-east and the May and June, following 4-5 months Fundación Produce de Yucatán— (December-April) of low precipitation involved developing a geographical (0-18 mm/month). Based on all of information system (GIS)-based model these parameters, risk assessment to predict the probability of whitefly maps for the Yucatán Peninsula were and begomovirus attack. produced. The assessments will allow recommendations to be issued According to the classification of regarding the most appropriate Toledo (1989), Mexico can be divided planting dates for those crops most into five major ecological zones. likely to be affected by B. tabaci, both Interestingly, the northernmost portion as a pest and vector of plant viruses. of the Yucatán Peninsula belongs to the Dry Tropics (Trópico Seco), the same ecological zone as the northern Pacific Socio-economic Analysis Plains, where the B. tabaci thrives. In this dry zone, the annual precipitation The results of the biological surveys ranges between 600 and 1500 mm and conducted during the course of the the mean annual temperature is over project show that whitefly-transmitted 20 °C. Although Yucatán is not one of viruses and B. tabaci affect several the country’s main horticultural crops of social and economic regions, in 1992 this peninsula importance throughout Mexico, with cultivated over 11,000 ha of Habanero the exception of the Central Highlands, and Jalapeño chilli pepper, watermelon where other whitefly species and tomato (listed in declining order of predominate. production area). However, the agriculture of the peninsula includes a In the Mexicali Valley, Baja diversity of other crops, including California and the San Luis Río species that can act as reproductive Colorado region of Sonora, B. tabaci hosts of B. tabaci such as soybean and attacked fields of cotton (damaging cotton of which over 500 ha of each are some 14,000 ha), summer melon grown. (causing the loss of 1500 ha) and sesame (Sesamum indicum L.) The GIS-based model for whitefly (damaging 7500 ha) in the 1991-92 and virus outbreaks drew on three season, causing nearly US$10,000,000 components of a database developed by in production losses (López, 1996). INIFAP: (1) a digital elevation model, Some 3000 ha of Serrano chilli and

183 Whiteflies and Whitefly-borne Viruses in the Tropics tomato crops were lost between 1992 common bean, chilli and eggplant and 1995 in the central region of San (Solanum melongena L.) (Ortíz et al., Luis Potosí State, because of damage 1996). Viruses transmitted by B. tabaci by B. tabaci and begomoviruses. Other have severely affected tomato plantings crops colonized by B. tabaci in this in the state of Guerrero. In order to region were common bean, cotton, harvest tomato, farmers have resorted tomatillo (Physalis ixocarpa B.), squash, to treating their fields twice a week melon, watermelon, cucumber with insecticides (Barajas, 1996). And (Cucumis sativus L. var. sativus), in the state of Yucatán, B. tabaci and sweetpotato (Ipomoea batatas [L.] Lam.) whitefly-borne begomoviruses have and sunflower (Helianthus annuus L.) affected horticultural crops such as hot (Hernández et al., 1996). peppers, tomato, squash, melon and watermelon since the 1980s (Díaz-Plaza B. tabaci caused severe yield losses et al., 1996). to various horticultural crops, especially melon, in the Comarca The social and economic impact of Lagunera, Durango, in 1993 (Morales these pest and disease outbreaks must et al., 1996). B. tabaci attacked melon, be enormous. In the mid-1990s, chilli and cotton in the state of horticultural and fruit crops occupied Coahuila, northern Mexico. The over 1 million hectares in Mexico. investigation that was conducted at the About 7 million tons of produce, mainly time showed that melon was an tomato, melon and watermelon were excellent host, cotton was a good host exported (principally to the United and chilli did not adequately support States, Canada, Japan and Brazil). The the reproduction of B. tabaci (Nava and rest, some 35 million tons of squash, Riley, 1996). Sonora has been one of melon, pepper, tomato, potato, the states in northern Mexico most eggplant and chillis, was absorbed by affected by infestations of B. tabaci on the national market for local various crops. An epidemiological study consumption, underlining the in the Valley of the Yaqui indicated that importance of these crops to the one of the main reproductive hosts of domestic economy. The production of the whitefly in the valley was soybean horticultural crops demands (Pacheco, 1996). The incidence of considerable human labour. For whitefly-transmitted begomoviruses in instance, tomato requires 271 work- common bean plantings in the valley of days per cropping cycle as compared El Fuerte, Sinaloa, caused yield losses with only 24 for maize (Zea mays L.). estimated at 20%-30% in the entire Production costs for maize were region, between February and June US$430 per hectare compared with 1994 (Salinas et al., 1996). B. tabaci US$3800 per hectare for tomato in has been increasing its populations on 1990. The marketing of horticultural cucumber in the valley of Culiacán, products is difficult because of their Sinaloa, since 1993, causing significant perishable nature and price instability. yield losses in over 9000 ha of this crop In 1988, over 50% of the farmers (Avilés and Valenzuela, 1996). producing horticultural products in Mexico had less than 5 ha of land In the state of Nayarit, the fall- (FIRA, 1997). Overall, the incidence of winter cropping cycle has been reduced whitefly-transmitted viruses affects a by 50% because of the direct and large portion of Mexico’s rural indirect damage caused by B. tabaci as population, many of whom can be a pest and vector of viruses in tomato, considered small-scale farmers.

184 Mexico

Strengthened Research Whereas the begomovirus situation Capacity on common bean has been partially controlled through the use of resistant In addition to its research activities, cultivars, the situation of other CINVESTAV offers training to visiting horticultural crops affected by whitefly- scientists from developing countries. transmitted begomoviruses such as During the course of this project, two tomato, pepper, chilli and melon is not Cuban scientists conducted research yet resolved in most regions. on Taino tomato mottle virus (TToMoV), Nevertheless, Mexico has made a whitefly-transmitted begomovirus considerable progress in implementing affecting tomato in Cuba (Ramos et al., integrated pest management (IPM) 1997). The project also contributed measures such as biological control, limited funds for research on viruses of physical barriers (micro-tunnels) and economic importance in horticultural other relevant cultural practices, and in crops to the Horticulture Program of developing the necessary knowledge INIFAP, and paved the way for base, including, for instance, virus continuing collaboration between characterization. Mexico is also at the INIFAP and the current Tropical forefront in using GIS technology for Whitefly Project. (The state of Yucatán more effective management of is the main target area for the current agricultural resources. Phase II of the project.) More research is needed to study the epidemiology of B. tabaci biotypes Current Status of and other whitefly species in Whitefly/Begomovirus agricultural regions with altitudes above Problems 1000 m, including those in the states of Oaxaca, Guerrero, Morelos, Mexico, San The whitefly/begomovirus situation in Luis Potosí, Guanajuato and Durango. Mexico is rather complex because of Suitable IPM measures should be the size and diversity of the country— rapidly implemented and evaluated to and because of the introduction or reduce dependence on chemical emergence of different whitefly- insecticides for whitefly control. transmitted viruses in both the north and the south. In the north, the intensive cropping systems for non- Acknowledgement traditional export crops share not only a common border but also common Project research and capacity-building whitefly/begomovirus problems with activities in Mexico were supported by the equally intensive agricultural funds made available through the systems of the south-western United TWF-IPM Project by the Danish Agency States. In the south, commercial and for Development Assistance (DANIDA). small-scale subsistence agriculture co- exist as an extension of the cropping systems found in Central America. References Thus, it is not surprising to find some North American begomoviruses in Ascencio-Ibáñez, J. T.; Díaz-Plaza, R.; Méndez-Lozano, J.; Monsalce- northern Mexico and some Central Fonnegra, J. I.; Arguello-Astorga, G.; American begomoviruses in southern Rivera-Bustamante, R. 1999. First Mexico. report of Tomato yellow leaf curl in Yucatán, Mexico. Plant Dis. 83:1178.

185 Whiteflies and Whitefly-borne Viruses in the Tropics

Avilés, M.; Valenzuela, G. 1996. FIRA (Fideicomisos Instituidos en Respuesta de diferentes materiales Relación con la Agricultura). 1997. de pepino al ataque de mosquita Rentabilidad de 15 cultivos anuales blanca (B. tabaci) en el Valle de financiados en el ciclo agrícola PV/ Culiacán, Sin. In: Resumen, Taller 1996. Boletín Informativo N. 293. de Moscas Blancas, Acapulco, MX. Banco de Mexico, MX. 34 p. p. 196. Flock, R. A.; Mayhew, D. E. 1981. Squash Barajas, R. 1996. Intervalos de aplicación leaf curl, a new disease of cucurbits de insecticida contra la mosquita in California. Plant Dis. 65:75-76. blanca Bemisia tabaci comparado con acolchado plástico en jitomate. Gallegos, H. M. L. 1978. Enchinamiento In: Resumen, Taller de Moscas del tomate (chino disease of Blancas, Acapulco, MX. p. 211. tomato). In: Enfermedades de cultivos en el estado de Sinaloa. Brown, J. K. 1989. A whitefly-transmitted Secretaría de Agricultura y geminivirus from peppers with tigre Recursos Hidráulicos (SARH), MX. disease. Plant Dis. 73:610. p. 119.

Brown, J. K. 1994. Current status of Garrido, E. R.; Gilbertson, R. L. 1998. Bemisia tabaci as a plant pest and First report of tomato mottle virus vector in agro-ecosystems geminivirus infecting tomatoes in worldwide. FAO Plant Protect. Bull. Yucatán, Mexico. Plant Dis. 82:592. 42:3-32. Hernández, V.; Obregón, J.; Contreras, J. Brown, J. K.; Chapman, M. A.; Nelson, M. L. 1996. Identificación de R. 1990. Bean calico mosaic, a new hospederos de mosquita blanca disease of common bean caused by Bemisia tabaci en la zona media a whitefly-transmitted geminivirus. potosina. In: Resumen, Taller de Plant Dis. 74:81. Moscas Blancas, Acapulco, MX. p. 177. Brown, J. K.; Idris, A. M.; Fletcher, D. C. 1993. Sinaloa tomato leaf curl virus, Hou, Y. M.; Garrido, E. R.; Stingley, S.; a newly described geminivirus of Guzmán, P.; Gilbertson, R. L. 1996. tomato and pepper in west coastal Characterization of the infectious Mexico. Plant Dis. 77:1262. clones of the Sinaloa strain of pepper huasteco geminivirus. Cancino, M.; Abouzid, A. M.; Morales, F. Phytopathology 86:S43. J.; Purcifull, D. E.; Polston, J. E.; Hiebert, E. 1995. Generation and INIFAP (Instituto Nacional de characterization of three Investigaciones Forestales Agrícolas monoclonal antibodies useful in y Pecuarias). 1995. Mosquita detecting and distinguishing Bean blanca en el noroeste de México. golden mosaic virus isolates. Memoria Científica no. 3. MX. 73 p. Phytopathology 85:484-490. López, G. H. 1974. Aumente sus Cárdenas, J. A.; Pérez, F.; Nieves, F. rendimientos en frijol en el Valle de 1996. Campaña contra la mosquita Culiacán. Circular no. 12, Centro blanca en México. In: Resumen, de Investigaciones Agrícolas del Taller de Moscas Blancas, Sureste (CIAS)-Instituto Nacional de Acapulco, MX. p. 167-169. Investigacíon Agraria (INIA)- Secretaría de Agricultura y Díaz-Plaza, R.; Avilés-Baeza, W.; Torres- Recursos Hidráulicos (SARH), MX. Pacheco, I.; Rivera-Bustamante, R. 1996. Geminivirus en la región hortícola de Yucatán, México. In: Resumen, Taller de Moscas Blancas, Acapulco, MX. p. 193.

186 Mexico

López, R. 1996. Experiencia sobre el Pacheco, J. J. 1996. Crecimiento control integral de la mosca blanca poblacional de la mosquita blanca Bemisia tabaci en el distrito de de la hoja plateada (Bemisia desarrollo rural 002, Rio Colorado, argentifolii) en el valle del Yaqui, Sonora. In: Resumen, Taller de Sonora. In: Resumen, Taller de Moscas Blancas, Acapulco, MX. Moscas Blancas, Acapulco, MX. p. 162. p. 181.

López-Salinas, E. 1994. Situación del Ramos, P. L.; Guerra, O.; Peral, R.; mosaico dorado del frijol en México- Oramas, P.; Guevara, R. G.; Rivera- Sur. In: Morales, F. J. (ed.). Mosaico Bustamante, R. 1997. Taino tomato dorado del frijol: Avances de mottle virus, a new bi-partite investigación. Programa geminivirus from Cuba. Plant Dis. Cooperativo Regional de Frijol para 81:1095. Centro América, México y el Caribe (PROFRIJOL)-Swiss Agency for Salinas, R.; Macías, C. J.; López, A. B. Development and Cooperation 1996. Caracterización de virus en (SDC), Centro Internacional de frijol a través de diferenciales bajo Agricultura Tropical (CIAT), Cali, condiciones de campo. In: CO. p. 28-33. Resumen, Taller de Moscas Blancas, Acapulco, MX. p. 192. Morales, Y.; Ramírez, M.; Pedroza, A. 1996. Dinámica poblacional de Sánchez, V.; Frías G; Aguirre, L.; Bravo, mosquita blanca Bemisia sp. en dos L.; Cotéz, L.; Quesada, J.; Ruíz, R; predios de melón con diferentes Hernández, M.; Rangel, U.; Garzón, prácticas de control. In: Resumen, J. 1996. Agro-ecology of a whitefly- Taller de Moscas Blancas, transmitted virus in an arid and Acapulco, MX. p. 178. isolated Serrano pepper growing region in Northeastern Mexico. Nava, C. U.; Riley, D. G. 1996. Desarrollo, Phytopathology 86:S117. sobrevivencia y fecundidad de Bemisia argentifolii (Homoptera: Stenger, D. C.; Duffus, J. E.; Villalon, B. Aleyrodidae) a diferentes 1990. Biological and genomic temperaturas y plantas hospederas. properties of a geminivirus isolated In: Resumen, Taller de Moscas from pepper. Phytopathology Blancas, Acapulco, MX. p. 179. 80:704-709.

Ortíz, M.; Arroyo, J. M.; Ramos, A.; Toledo, V. M. 1989. La producción rural Gonzáles, J. 1996. Especies de en México: Alternativas ecológicas. mosquitas blancas, sus enemigos Ed. Fundación Universo 21, MX. naturales y hospederos en el estado 76 p. de Nayarit. In: Resumen, Taller de Moscas Blancas, Acapulco, MX. p. 204.

187 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.3 Guatemala

Francisco Morales*, Abelardo Viana**, Margarita Palmieri***, Mónica Orozco*** and René Ruanoψ

Introduction creating complex and fragile cropping systems. Figure 1 shows the main agricultural regions affected by Geographical context whitefly-transmitted begomoviruses. Of the seven countries that comprise the isthmus of Central America, Guatemala is physically and culturally the most diverse. The highlands of the southern half of Guatemala descend southward to the Pacific Coast to form Guatemala the fertile “Piedmont” or “Coffee Belt” (500-1500 m above sea level) and the hot, grass- and forest-covered lowlands known as “The Coast” (0-500 m). The Baja Verapaz northern regions of Guatemala are Zacapa tropical lowlands covered by rain forest El Progreso and scattered savannahs, which are Jalapa only sparsely populated (West and Santa Jutiapa Augelli, 1977). However, it is the Escuintla Rosa south-eastern region of the country (“El Oriente”), formed by low- and mid- Figure 1. The main agricultural regions altitude (200-1000 m) valleys, which affected by whitefly-transmitted begomoviruses, Guatemala. has suffered the worst attacks from the whitefly Bemisia tabaci (Gennadius) and different begomoviruses transmitted by this whitefly species. The emergence of Bemisia This region has drastically changed its tabaci as a pest and virus traditional agricultural environment to vector join the boom in export crops, thus The civil war in the United States in the 1860s created an increased demand for cotton (Gossypium hirsutum L.). Soon, * Centro Internacional de Agricultura Tropical cotton produced in the Pacific lowlands (CIAT), Cali, Colombia. of the country comprised 20% of ** Proyecto Regional de Frijol para Centro América, México y el Caribe (PROFRIJOL), Guatemala’s exports. This species was Guatemala City, Guatemala. probably one of the first commercially *** Universidad del Valle, Guatemala City, cultivated hosts of the whitefly Guatemala. ψ Instituto de Ciencia y Tecnología Agrícolas B. tabaci. A second boom of cotton (ICTA), Guatemala City, Guatemala. production took place in the 1950s and

188 Guatemala

1960s, and at this time the first viruses Programa Cooperativo Regional de transmitted by whiteflies were observed Frijol para Centro América, México y el in Guatemala and neighbouring Caribe (PROFRIJOL), initiated by the countries (Gill, 1994). Centro InternacionalSonora de Agricultura Tropical (CIAT) to address the bean Tobacco (Nicotiana tabacum L.) is golden yellow mosaic problem. This another popular export crop that project, initiated in 1978 with support continues to be grown in Guatemala. from the United Nations Development The cultivation of tobacco in south- Programme (UNDP) and subsequently eastern Guatemala has been blamed funded (from 1980) by the Swiss also for increasing the population of Development Cooperation (SDC), is the virus vector B. tabaci in the 1970s, credited with the development of thus provoking the first outbreaks of several BGYMV-resistant (DOR) lines Bean golden yellow mosaic virus that have subsequently been widely (BGYMV) in common bean (Phaseolus adopted as commercial varieties in vulgaris L.) in that region of Guatemala Mexico, Central America, the (Morales, 1994). Caribbean and South America.

In 1974, the Horticultural In Guatemala, the project began Research Division of the Instituto de work under the auspices of ICTA in the Ciencia y Tecnología Agrícolas (ICTA) south-eastern Monjas Valley, about initiated extension work in the Zacapa 960 m above sea level. Until the 1970s, Department with a view to producing this valley had been planted to export crops such as melon (Cucumis traditional crops, mainly common bean melo L.). At that time, the main and maize (Zea mays L.). In the 1980s, production problems identified were horticultural crops such as tomato and outdated agronomic practices, broccoli (Brassica oleracea L. var. italica inadequate cultivars and the poor Plenck) made their appearance in the quality of the produce, which could valley. Other non-traditional crops— only meet the relatively low standards pepper (Capsicum spp. L.), cucumber of the local markets. A similar industry (Cucumis sativus L. var. sativus) and to produce tomato (Lycopersicon more recently watermelon (Citrullus esculentum Mill.) was also initiated in lanatus [Thunb.] Matsum. & Nakai) and the late 1970s in south-eastern grape (Vitis vinifera L.)—have also Guatemala (Gaytán, 1979). These crops found a niche in this valley over the provided better reproductive hosts for latter part of the 1990s. Tobacco is a B. tabaci than either common bean or large-scale commercial crop that was wild plant species. The populations of first planted in the valley during the the vector soon soared and then new early 1970s; its production collapsed in begomoviruses emerged to attack the the late 1980s, only to make a new crops. The arrival of the B biotype comeback in recent years. of B. tabaci, a more aggressive and polyphagous variant of the original BGYMV, appearing in the mid- species, further aggravated yield losses, 1970s, was the first whitefly- with the new whitefly acting as both a transmitted virus to cause severe yield pest and a vector of damaging plant losses in the Monjas Valley viruses (Dardón, 1992). (Rodríguez, 1994). Tobacco was probably the most important The first collaborative project that reproductive host of B. tabaci. The local set out to tackle the new and growing common bean cultivars, Pecho Amarillo threat of whitefly-borne viruses was the and Rabia de Gato, soon succumbed to

189 Whiteflies and Whitefly-borne Viruses in the Tropics the disease. Fortunately, the Table 1. Antigenic properties of common bean PROFRIJOL project began to show begomoviruses from Guatemala. results and in due course released the Isolate Locality MAB-BSa MAB-GAb first disease-resistant breeding lines: 1 Monjas + + ICTA-Quetzal, ICTA-Jutiapán and 2 Tecpán + - ICTA-Tamazulapa (Rodríguez, 1994). 3 Cuyuta + - The causal agent, initially believed to 4 Monjas + - be Bean golden mosaic virus (BGMV), 5 Monjas + - was characterized at the molecular 6 San Jerónimo + - level, thanks to collaborative research 7 Quiaté + - financed by the United States Agency a. MAB-BS, a broad spectrum monoclonal for International Development and antibody used to detect bi-partite carried out at the University of begomoviruses. Wisconsin and at CIAT. The b. MAB-GA, a monoclonal antibody used to Guatemalan virus in the early 1990s detect the original Middle American isolates of Bean golden yellow mosaic virus- was shown to be a distinct species, Guatemala. different from the Brazilian isolates of BGMV (Faria et al., 1994) but with a considerable degree of homology and with the original Middle-American causing a very similar disease; it was BGYMV isolates. The result that MAB- therefore given the name Bean golden GA, raised against the original isolate yellow mosaic virus. Maize is not a host from Monjas, does not react with isolates of the whitefly B. tabaci and is not subsequently collected from this area attacked by these particular suggests that some change in the begomoviruses, although maize in antigenic properties of the virus has Africa is infected by different occurred or that a new isolate has been geminiviruses (Geminiviridae: introduced and become established in Mastrevirus) transmitted by the area. Begomoviruses are known to leafhoppers. adapt to different host plants and vector species. It has been suggested that the changes observed in the antigenic Advances in Biological reaction of the Monjas isolates may be Research associated with the displacement of the A biotype of B. tabaci by the B biotype, Further research, undertaken in which is believed to have occurred in the collaboration with the University of 1990s. Florida, was intended to produce monoclonal antibodies (MABs) against Characterization of whiteflies different isolates of BGYMV, using the collected from sites across Guatemala Guatemalan (MAB-GA) isolate of confirms that the B biotype has displaced BGYMV. At the same time, the project the A biotype almost completely (Table 2). sought to produce a broad-spectrum This would be consistent with the monoclonal antibody (MAB-BS), hypothesis that the antigenic changes in capable of detecting most whitefly- the virus detected with the specific MAB transmitted begomoviruses (Cancino may be related to the presence of a new et al., 1995). The results shown in vector. To explore this possibility further, Table 1 indicate that these objectives the coat protein of a recent isolate of were realized, in as much as MAB-BS BGYMV from the locality of Monjas was has been able to detect all bi-partite partially sequenced and compared with begomoviruses encountered to date in the original isolate. Table 3 shows the Guatemala, but MAB-GA reacts only comparative results obtained.

190 Guatemala

Table 2. Results of whitefly surveys in different agricultural regions of Guatemala. Data are numbers of samples that included Bemisia tabaci biotype A (B.t.-A), B. tabaci biotype B (B.t.-B) or other species of whitefly (other spp.).

Crop Samples Localitya B.t.-A B.t.-B Other spp.

Melon 12 Cuyuta-Jutiapa 0 12 0 Melon 6 Río Hondo-Zacapa 0 6 0 Watermelon 10 La Fragua-Zacapa 0 10 0 Watermelon 22 L.de Retana-Jutiapa 0 2 20 Cucumber 17 Usumatlán-Zacapa 0 17 0 Cucumber 10 AMCO-Jutiapa 0 10 0 Cucumber 10 S. Agustín-Jutiapa 0 10 0 Okra 7 La Fragua-Zacapa 0 7 0 Tobacco 2 Santa Cruz-Jutiapa 0 2 0 Tomato 2 Miramar-Jutiapa 2 0 0 Tomato 1 AMCO-Jutiapa 0 1 0 Eggplant 10 San Agustín-Jutiapa 0 10 0 a. AMCO-Jutiapa, an experiment station.

Table 3. Comparative analysis of partial coat crops assayed, with the exception of protein sequence homologies (%) cucumber. Tobacco is also beginning to among viruses associated with Bean golden yellow mosaic virus, be attacked by begomoviruses, as comparing the isolate GA-2, collected production of this crop increases once in Guatemala during the project, with more. Tobacco, however, was observed GA-1, the original isolate, and isolates from the Dominican Republic to be principally affected by Tobacco (DR), Puerto Rico (PR) and Bean mosaic virus (TMV) (Tobamovirus), a golden mosaic virus- Brazil (BR). highly infectious virus that is spread by Virus GA-2 GA-1 DR PR BR workers entering the field. A similar GA-2 100 92.7 92.3 90.0 76.0 virus, probably Tomato mosaic virus, was also found infecting tomato in the south-eastern region of Guatemala, These results show a divergence in although the two viruses may be coat protein nucleotide sequence of ca. strains of TMV adapted to each crop. 7.3% between the original and recent Guatemalan isolates. This difference is The presence of Pepper golden significant, considering that when mosaic virus in pepper was confirmed sequences of two isolates diverge by in samples taken from the locality of more than 10% they are normally Laguna de Retana. The collaborating regarded as distinct species—as in the Plant Pathology Laboratory of the case of the difference observed University of Wisconsin had detected between BGYMV-GA and BGMV-BR this virus already in Guatemala (see Table 3). (Douglas Maxwell, personal communication, 2003). Table 4 shows the results of serological and other complementary diagnostic tests carried out on plant Socio-economic Analysis samples taken from selected crops in whitefly-affected regions in Guatemala. Over half of Guatemala’s total These results demonstrate the population of some 12 million people presence of whitefly-transmitted are of pure Amerindian stock. This begomoviruses in most horticultural sector of the population is concentrated

191 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 4. Results of assays carried out on selected plant samples from Guatemala.

Sample Locality Plant Reactiona EMb MAB-BS PTY1

1 Zacapa Melon + - nt 2 Zacapa Melon - + nt 3 Monjas Tobacco + nt nt 4 Monjas Tobacco + nt TMV 5 El Ovejero Tobacco - nt TMV 6 L. de Retana Tomato + nt nt 7 L. de Retana Tomato - - nt 8 L. de Retana Tomato + nt nt 9 Zacapa Tomato - - nt 10 Las Conchas Tomato - nt TMV 11 Las Conchas Tomato - nt TMV 12 Las Flores Tomato + nt nt 13 L. de Retana Tomato + nt nt 14 L. de Retana Tomato + nt nt 15 Plan de la Cruz Tomato + nt nt 16 Plan de la Cruz Tomato + nt nt 17 Monjas Tomato + nt nt 18 Monjas Tomato + nt nt 19 Monjas Tomato + nt nt 20 L. de Retana Pepper + nt nt 21 L. de Retana Pepper - nt - 22 L. de Retana Pepper + nt nt 23 L. de Retana Cucumber - nt - 24 L. de Retana Cucumber - nt - a. Positive or negative reactions to a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses (MAB-BS); reactions to a potyvirus-specific monoclonal antibody (PTY1); and nt, no tests for these samples. b. Detection of Tobacco mosaic virus (TMV) by electron microscopy (EM); and nt, no tests for these samples. in the highlands, north and west of located at an altitude of 960 m and has Guatemala City, as well as in the an average annual precipitation of region of Petén. South-eastern 900 mm, most of which falls between Guatemala is the region with the lowest May and October. The mean proportion of native Americans. The temperature is 23.7 °C and the mean presence of mestizos or ladinos in the relative humidity is 80%. Ecologically, Oriente of Guatemala explains the more the Monjas Valley is classified both as dynamic and complex cropping dry subtropical forest and temperate systems in these drier lowlands, subtropical humid forest. Most of the including the prevalence of non- agricultural land is private property, traditional export crops. although it may have been acquired by “right” and not through a commercial A special study was conducted in transaction. the municipality of Monjas, Jalapa Department, where Amerindians still Table 5 shows the existing make up over 70% of the total cropping systems in the Monjas Valley. population. The Monjas Valley is Maize and common bean are evidently

192 Guatemala

Table 5. Cropping systems in the Monjas Table 6. Growers’ perception of whitefly Valley, Jalapa, Guatemala. problems and growers’ use of pesticides in selected cropping Crop Years of Irrigation Rain-fed systems of the Monjas Valley, Jalapa. planting Crop Whitefly Pesticides used Common bean >50 - + problem (no.) Maize >50 - + Common bean Yes 8 Tobacco 25 + + Maize No 1 Tomato 15 + - Tobacco Yes 9 Broccoli 10 + - Tomato Yes 11 Onion 10 - + Broccoli Yes 1 Pepper 7 + + Onion No - Cucumber 5 + - Pepper Yes 5 Sweet corn 3 + - Cucumber Yes 10 Grape 3 + - Sweet corn No 1 Grape No - the most traditional crops in the valley. Tobacco has been planted for over 20 years as a commercial crop but crops listed above, and the number of almost disappeared in the late 1970s pesticides used in each crop. The because of market problems, coming results clearly illustrate that pesticide back into favour only in the late 1990s. abuse occurs in most of the crops Onion (Allium cepa L.) is another crop affected by whiteflies. In the case of that has been planted in the valley for tomato, only five of the 11 pesticides over 10 years but it has never become used are specific to the control of prevalent. Broccoli has been a very whitefly; the remaining six pesticides popular crop in the valley in the past are applied as “repellents” in the belief decade and pepper has been that the “smell” of these products commercially cultivated during the past repels the whitefly. The low use of 7 years. Cucumber has been planted insecticides on broccoli is interesting commercially for the past 5 years and and responds to demand from foreign watermelon for the past 2 years, buyers (many of whom only allow bio- although it has been planted for local pesticides to be used). consumption for decades in the valley. Finally, sweet corn (Zea mays L. subsp. In general terms, whiteflies are the mays) and grape have been introduced main pest on common bean, tomato, in the past 3 years. broccoli and tomato. Other pests such as white grubs and lepidopterous Table 5 also shows the association larvae also cause significant damage to between commercial crops and the use tobacco, tomato and sweet corn. Slugs of irrigation in the Monjas Valley. and chrysomelids are perceived as Subsistence crops such as maize and important pests of common bean in the common bean are planted during the valley. Diamond-back moth, Plutella rainy season. The use of irrigation xylostella (Linnaeus), is also a pest of demonstrates the intention of local broccoli; and the corn ear worm, growers to minimize risk in the case of Helicoverpa zea (Boddie) causes severe high-value crops. damage to maize and sweet corn. The existence of insect pests other than Table 6 shows the perception that whiteflies in the various crops grown in growers have of the whitefly B. tabaci, the Monjas Valley hampers efforts to both as a pest and virus vector in the reduce the use of chemical pesticides

193 Whiteflies and Whitefly-borne Viruses in the Tropics against whiteflies. In fact, most growers misinterpreted. Most growers of non-traditional export crops use responded that they were always ready “cocktails”, or “bombs”, as pesticide to control whiteflies in seedbeds, after mixtures are locally called. transplanting and during the vegetative and early reproductive phases of the Whereas growers recognize crop. In fact, some growers indirectly B. tabaci as a pest, they do not fully answered the question, associating understand its role as a vector of mild winters and an abundance of viruses. However, they readily associate alternative crops with higher whitefly the bright yellowing shown by infected populations and crop damage. common bean plants with a disease (bean golden yellow mosaic). In the case of tomato, growers call symptoms Strengthened Research expressed by begomovirus-infected Capacity plants acolochamiento (curling). In the case of tobacco, growers recognize the The Tropical Whitefly Integrated Pest severe malformation and stunting Management (TWF-IPM) Project symptoms shown by virus-infected financed two case studies conducted by plants. However, they cannot ICTA agronomists, in the departments distinguish between whitefly- of Jalapa, Jutiapa and Baja Verapaz. transmitted begomoviruses and TMV. These departments used to be traditional agricultural areas, devoted According to most growers, to the planting of food staples such as whiteflies are a regular pest, year after maize and common bean. The boom in year, particularly during the summer export crops, however, rapidly time because of the higher displaced traditional crops and temperatures and relative humidity. drastically changed the socio-economic Growers also perceive the complex structure of the rural communities cropping systems in the valley as a found in these departments. The negative factor because they claim that studies conducted in these regions these crops represent more food provided information on the positive sources for the whitefly. Finally, aspects (potential higher income) and growers blame ratoon crops for the negative consequences (new biotic survival of whiteflies between summer problems and pesticide abuse) of non- and winter seasons. Growers believe in traditional export crops. The first study the existence of whitefly-resistant or was conducted in 15 rural more generally pest- and disease- communities, located in three resistant crops, particularly tomato, municipalities of the departments of tobacco and common bean. However, Jalapa and Jutiapa. The second case the “resistance” of tomato and tobacco study was conducted in five rural seems to be associated with the communities in the municipalities of implementation of more successful San Jerónimo, Rabinal and San Miguel chemical control practices in the valley. Chicaj. This study helped to develop In reality, the case of common bean is non-chemical strategies for whitefly a good example of genetically controlled control and generated information of virus resistance obtained through a value to ICTA in its attempts to adopt well-designed breeding programme. suitable research policies to assist small-scale farmers. A general question on the ability of growers to predict seasons of high or Although training in advanced low whitefly incidence was generally laboratory techniques was not

194 Guatemala contemplated in this preliminary phase economic knowledge base on which of the project, it became apparent that research and implementation of a number of agricultural scientists in agricultural policy are founded. The Central America, Mexico and the research sector of Guatemala was also Caribbean have adequate laboratory strengthened through the training of facilities to conduct such work. These scientists from advanced research scientists are becoming increasingly laboratories in the country in reluctant to act simply as “guides” or techniques used for the identification of “providers” of biological samples to whitefly biotypes and the foreign scientists from advanced begomoviruses that affect food and research laboratories that often visit export crops of social and economic developing countries in search of importance. interesting scientific materials. Although the aim of this project was to conduct research in a truly Current Status of collaborative manner, in which national Whitefly/Begomovirus scientists have preferential access to Problems the results of all research undertaken, the possibility of training national The whitefly/begomovirus problems in scientists in advanced molecular Guatemala are concentrated in the techniques became a need. The two south-eastern agricultural regions of research areas in which advanced the country (Figure 1). This region techniques are being implemented in contains most of the country’s mid- this project are in the molecular altitude valleys (at 500 to 1000 m identification of whitefly biotypes and altitude) with an average annual of the begomoviruses they transmit. rainfall below 1500 mm, and temperatures between 20 and 25 ºC; In Guatemala, the identification of these valleys are also characterized by begomoviruses is being done at the highly diverse cropping systems, University of San Carlos with the including irrigation. All of these factors support of the Plant Pathology favour the reproduction of B. tabaci. Laboratory of the University of Moreover, the concentration of non- Wisconsin. The molecular traditional export crops in this region is characterization of B. tabaci biotypes is associated with the intensive use of being carried out at the University of pesticides, regularly used to protect El Valle, Guatemala City, which has these high-value crops. Pesticide abuse adequate facilities to conduct research is another predisposing factor for the in the area of molecular biology but did establishment of B. tabaci as a pest. not have the technology to identify Any crop planted in the Pacific whitefly biotypes. Consequently, a lowlands will be severely attacked by research assistant from the relevant whitefly-transmitted begomoviruses laboratory received 1 week of intensive particularly because of the higher training at CIAT on the molecular temperatures, which increase disease characterization of B. tabaci biotypes severity. and was able to carry out this work successfully upon return to The current begomovirus situation Guatemala. is becoming more complex as new whitefly-transmitted viruses are Thus, the project helped national introduced from neighbouring scientists to conduct surveys that countries, or evolve from the improved both the biological and socio- indigenous virus population, or arise

195 Whiteflies and Whitefly-borne Viruses in the Tropics through genetic recombination between Dardón, D. E. 1992. Las moscas blancas the existing begomoviruses that affect en Guatemala. In: Hilje, L.; Arboleda, the diverse crops such as common O. (eds.). Las moscas blancas en América Central y el Caribe. bean, tomato, pepper, squash Memorias del Taller Centroamericano (Cucurbita spp. L.), melon and tobacco y del Caribe sobre moscas blancas, that are grown in this area. Only in the Turrialba, CR. p. 38-41. case of common bean has there been a sustained effort on the part of the Faria, J. C.; Gilbertson, R. L.; Hanson, S. national program and CIAT (with Swiss F.; Morales, F. J.; Ahlquist, P.; funding) to develop and release virus- Loniello, A. O.; Maxwell, D. P. 1994. Bean golden mosaic geminivirus resistant cultivars. For the remaining type II isolates from the Dominican crops, the emphasis has been on Republic and Guatemala: Nucleotide chemical control, although large agro- sequences, infectious export companies have adopted some pseudorecombinants, and IPM packages. South-eastern phylogenetic relationships. Guatemala provides an ideal site to Phytopathology 84:321-329. conduct a case study with small-scale Gaytán, M. A. 1979. Proyecto de farmers on the implementation of exportación de melones en el Valle practical IPM measures to combat B. de Zacapa. XXV Reunión annual tabaci and the begomoviruses that this Programa Cooperativo species transmits to food and Centroamericano para el industrial crops in this region. Mejoramiento de Cultivos Alimenticios (PCCMCA), Tegucigalpa, HN.

Acknowledgements Gill, A. 1994. Problemática del complejo mosca blanca-virus en algodón en The authors are grateful to the Centroamérica. In: Mata, M.; Dardón, following technical staff of ICTA who D. E.; Salguero, V. E. (eds.). Biología assisted in the research: Carlos Paiz y manejo del complejo mosca blanca- García, Edgardo Carrillo, Edin Palma, virosis. Memorias III Taller Iván Esquivel, Viany Gillespie, José Centroamericano y del Caribe sobre mosca blanca, Antigua, GM. Luis Ordóñez and T. Mairor R. Osorio. p. 23-38. The work was supported by a grant from the Danish Agency for Morales, F. J. 1994. Bean golden mosaic: Development Assistance (DANIDA), via Research advances. Centro the TWF-IPM Project. Internacional de Agricultura Tropical (CIAT), Cali, CO. 193 p.

Rodríguez, R. 1994. Situación actual del References mosaico dorado del frijol en Guatemala. In: Morales, F. J. (ed.). Cancino, M.; Abouzid, A. M.; Morales, F. Bean golden mosaic. 1994 research J.; Purcifull, D. E.; Polston, J. E.; advances. Programa Cooperativo Hiebert, E. 1995. Generation and Regional de Frijol para Centro characterization of three América, México y el Caribe monoclonal antibodies useful in (PROFRIJOL)-Swiss Development Bean detecting and distinguishing Cooperation (SDC). Centro golden mosaic virus isolates. Internacional de Agricultura Tropical Phytopathology 85:484-490. (CIAT), Cali, CO. p. 34-39.

West, R. C.; Augelli, J. P. 1977. Middle America: Its lands and peoples. Second edition. Prentice-Hall, NJ, USA. 494 p.

196 El Salvador

CHAPTER 3.4 El Salvador

Francisco Morales*, Carlos Pérez**, Priscila Henríquez**, Reina de Serrano**, Leopoldo Serrano*** and Claudio Nunesψ

Introduction officinarum L.) became an important cash crop at the turn of the century and a third of the sugar produced is Geographical context exported to the United States. Rice El Salvador lies entirely within a (Oryza sativa L.) is another important volcanic region but most of its territory crop as well as some cucurbits such as is formed by mid to low altitude valleys watermelon (Citrullus lanatus [Thunb.] (up to 1000 m). About 40% of El Matsum. & Nakai) (Pastor, 1988). More Salvador consists of agricultural land— recently, the production of melon the highest proportion in Latin (Cucumis melo L.), pepper (Capsicum America, where it is the smallest spp. L.) and tomato (Lycopersicon country but the most densely esculentum Mill.) has greatly increased populated. Given the shortage of land, because of their relatively high market the higher slopes are dedicated to value; even so, some of these products crops such as coffee (Coffea arabica L.). continue to be imported because the Farmers make up over 50% of the local market absorbs most of the population, primarily producing country’s produce. Common bean is subsistence crops such as maize (Zea widely grown in El Salvador, occupying mays L.) and common bean (Phaseolus about 75,000 ha (1995). However, total vulgaris L.). common bean production does not satisfy internal demand and, in 1994- Cotton (Gossypium hirsutum L.), 95, El Salvador imported 14,600 tons undoubtedly the second most of common bean (Parada and Pérez, important commercial crop in El 1996). Salvador in past decades (for export to Japan), is mostly found in the Pacific Because of the relatively small size lowlands. Sugarcane (Saccharum of the country, there are few ecologically distinct areas other than those defined by altitude, where the whitefly Bemisia tabaci (Gennadius) cannot thrive. Hence, the whitefly/ * Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. begomovirus problem is distributed ** Centro Nacional de Tecnificación Agrícola fairly evenly throughout the country. (CENTA), El Salvador. Figure 1 shows the main agricultural *** Universidad del Salvador. ψ Universidad Técnica Latinoamericana, regions affected by whitefly-transmitted El Salvador. begomoviruses.

197 Whiteflies and Whitefly-borne Viruses in the Tropics

maize and common bean were largely Santa Ana displaced by high-value horticultural El Salvador crops (e.g., tomato, pepper and Ahuachapan cucurbits). However,Sonora these vegetable Cuscatlan La crops have been eliminated gradually Libertad from the valley because of the damage Sonsonate San Vicente caused by whitefly-transmitted viruses.

Figure 1. The main agricultural regions Advances in Biological affected by whitefly-transmitted Research begomoviruses, El Salvador. The first generation of black-seeded, BGYMV-resistant common bean The emergence of Bemisia varieties produced in Guatemala in the tabaci as a pest and virus late 1970s made no impact on the vector control of BGYMV in El Salvador, Granillo and co-workers (1974) first because Salvadorans prefer bright red- reported B. tabaci as a vector of plant coloured beans. It was not until 1992, viruses in El Salvador. The first virus when the first red-seeded bean varieties and whitefly outbreaks occurred in the were developed and released, that bean Pacific coastal area, affecting cotton, could be cultivated again in these mid- kenaf (Hibiscus cannabinus L.) and altitude valleys. The BGYMV isolates common bean. As in the case of from El Salvador are very invasive in Guatemala, we find here a close susceptible common bean cultivars association between cotton and the such as Rojo de Seda, the preferred whitefly B. tabaci. Among the biotic common bean cultivar in the country. problems that limit common bean Nevertheless, the BGYMV isolates from production in the country, the main El Salvador were shown to be closely pest and disease problems are B. tabaci related to the Middle American BGYMV and Bean golden yellow mosaic virus isolates from Guatemala and Puerto (BGYMV). Zaumeyer and Smith (1964) Rico (Table 1). first reported this virus in El Salvador. BGYMV was more commonly observed in the Pacific lowlands, reaching up to Table 1. Serological reactions of selected begomoviruses from El Salvador. 100% incidence in most affected common bean plantings in the BGYMVa Year MAB-BSb MAB-GAc departments of La Libertad, La Paz, San isolate collected Vicente and Usulutan. BGYMV has Guatemala 1992 + + reached most of the adjacent mid- El Salvador 1992 + + altitude valleys in the departments of Puerto Rico 1992 + + La Libertad and Ahuchapan, and it can El Salvador 1998 + - be found now wherever common bean is El Salvador 1999 + - grown in the country. The introduction El Salvador 1999 + - of new crops, particularly horticultural a. BGYMV, Bean golden yellow mosaic virus. crops, further complicated the dynamics b. MAB-BS, a broad spectrum monoclonal of whitefly populations in El Salvador. antibody used to detect bi-partite begomoviruses. The Zapotitán Valley (La Libertad) is a c. MAB-GA, a monoclonal antibody used to good example of a traditional detect the original Middle American isolates agricultural area where crops such as of BGYMV-Guatemala.

198 El Salvador

During the course of this project, The next step was to investigate individual virus samples were collected the proportion of whitefly-transmitted from common bean plants showing viruses in the different crops grown in typical BGYMV symptoms. Table 1 whitefly-affected agricultural regions. shows the results of the serological Table 3 shows the results from this tests, with the broad spectrum and investigation. Basically, it is evident specific (Guatemalan) monoclonal that whiteflies are not the only group of antibodies developed for BGYMV, for vectors transmitting viruses to BGYMV isolates collected in 1992, 1998 horticultural crops in El Salvador. and 1999. The results show that, as of Many of the samples tested also 1998, there have been changes in the contained potyviruses transmitted by coat protein composition of BGYMV aphids. It is interesting to note that the isolates in El Salvador (no reaction with cucurbit locally called pipián (Cucurbita the specific MAB-GA antiserum). argyrosperma C. Huber subsp. argyosperma) had not been recorded In order to determine the current previously in Central America as a host composition of whitefly species/biotypes of whitefly-transmitted viruses. For the in El Salvador, a collection of immature four cucurbit species assayed in the whiteflies found on economically above survey, aphid-transmitted important crops affected by potyviruses appear as the main viral begomoviruses and/or B. tabaci was problem, with the exception of pipián. undertaken in selected agricultural For common bean, BGYMV was departments of the country. The results present but the relatively low number obtained (Table 2) show an interesting of samples collected probably reflects situation of mixed populations of the displacement of this food legume biotype A and B of B. tabaci, biotype B from areas traditionally growing being predominantly associated with common bean. The results obtained for cucurbits. tomato and pepper suggest a complex

Table 2. Results of whitefly biotype surveys in different agricultural regions of El Salvador.

Cropa Sample Department B.t.-Ab B.t.-Bb Other

Tomato 1 Chalatenango 9 - 1 Chilli 2 Chalatenango 4 5 1 Watermelon 3 La Libertad 7 - 3 Loroco 4 La Libertad - 3 7 Chilli 5 La Libertad 7 - 3 Ayote 6 Santa Ana 6 - 4 Watermelon 7 La Paz 8 1 1 Melon 8 La Paz 2 7 1 Pipián 9 La Paz 4 - 6 Tomato 10 Chalatenango 5 - 5 Tomato 11 Chalatenango 18 - 2 Watermelon 12 Santa Ana 9 - 1 Tomato 13 Ahuachapán 8 - 2 Cucumber 14 Ahuachapán 4 3 3 Ayote 15 Ahuachapán 4 4 2 Watermelon 16 Santa Ana 2 6 2 a. Loroco (Fernaldia pandurata [A. DC.] Woodson), ayote (Cucurbita moschata Duchesne), pipián (Cucurbita argyrosperma C. Huber subsp. argyrosperma). b. Bemisia tabaci A and B biotypes.

199 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 3. Serological analyses of plants showing virus-like symptoms associated with the presence of whiteflies in El Salvador.

Samples Planta Begomovirus Potyvirus Other None

45 Pipián 25 11 0 9 9 Watermelon 0 9 1 - 11 Cucumber 0 7 0 4 6 Ayote 1 2 0 3 6 Common bean 2 0 0 4 38 Tomato 0 2 2 36 19 Pepper 0 4 0 0 1 Okra 0 0 0 0 1 Loroco 0 1 CMVb 0 1 Chilipuca (lima bean) 1 0 0 0 a. Pipián (Cucurbita argyrosperma C. Huber subsp. argyrosperma), ayote (Cucurbita moschata Duchesne), loroco (Fernaldia pandurata [A. DC.] Woodson), chilipuca (Phaseolus lunatus L. var. lunatus). b. CMV, cucumovirus. virus problem in these crops. In past coat protein level (AV1) probably reflect tests, one pepper sample and one minor changes in certain aminoacids tomato sample collected earlier in the as a result of the introduction of a new Zapotitán Valley had tested positive for whitefly biotype. Differences at the the presence of begomoviruses. replicase level (AC1) may represent an evolutionary trend. The significant A BGYMV isolate from San Andrés, differences between the Middle El Salvador, was selected for partial American BGYMV species and the sequencing to determine the degree of Brazilian BGMV species reflect their molecular divergence with respect to different origins or divergent BGYMV isolates characterized in the evolutionary paths. early 1990s. Table 4 shows the comparative results obtained with two A visit to one of the most partial regions of the BGYMV genome: traditional agricultural areas of El the viral replicase (AC1) and the coat Salvador, the Zapotitán Valley, showed protein (AV1). As can be observed, the the profound changes that have Salvadoran BGYMV-ES isolate has not occurred in the cropping systems of changed drastically over the past this country. This irrigation district, decade and it can still be considered an thanks to its proximity to the capital, isolate of the Middle American BGYMV San Salvador, was the breadbasket of species (sequence homologies above El Salvador, producing most of the 90%). The differences observed at the basic food crops demanded by the local

Table 4. Partial sequence homology (%) between a Bean golden yellow mosaic virus (BGYMV) isolate from El Salvador (ES) and related BGYMV/Bean golden mosaic virus (BGMV) isolates characterized in the early 1990s.

ORFa Isolatesb BGYMV-DR BGYMV-PR BGYMV-GA BGMV-BR BGYMV-ES

AC1 90.5 88.8 91.7 69.4 100 AV1 94.3 91.5 93.8 78.2 100 a. ORF, open reading frame; AC1, viral replicase gene; and AV1, coat protein gene. b. DR, Dominican Republic; PR, Puerto Rico; GA, Guatemala; and BR, Brazil.

200 El Salvador markets: common bean, maize, etc. In Socio-economic Analysis the past decade, this valley has experienced drastic changes in its A case study was conducted in agriculture to meet the increasing El Salvador, in the San Vicente demand for higher value crops, Department. The study was entitled particularly horticultural crops such “Farmers’ perception of the whitefly as tomato, pepper and cucurbits. A B. tabaci as an agricultural pest: recent survey of the Zapotitán Valley management and environmental revealed the predominance of factors affecting the incidence of the horticultural crops and the high whitefly in the department of San incidence of viral diseases affecting Vicente”. This department has an them. Eight samples (Table 5) were agricultural area of 55,340 ha, taken selectively from different fields including 20,346 ha planted to basic visited in the valley, all from plants grains—maize, sorghum (Sorghum showing symptoms usually associated bicolor [L.] Moench), common bean, with whitefly-borne viruses. sesame (Sesamum indicum L.) and soybean (Glycine max [L.] Merr.)— This preliminary survey, 4015 ha are cultivated to coffee, conducted in order to define priorities 2844 ha are in sugarcane and the rest in Phase II of the Tropical Whitefly in horticultural and fruit crops. Among Project, shows that other viruses are the horticultural crops grown here, present in the valley, besides whitefly- ayote (Cucurbita moschata Duchesne), transmitted begomoviruses. The pipián, sweet pepper (Capsicum viruses detected by electron annuum L. var. annuum), hot pepper microscopy and with the use of the (Capsicum spp. L.), tomato, cucumber, monoclonal antibody PTY1 are melon and watermelon are the most potyviruses transmitted by aphids and important. predominantly were found affecting pepper and cucumber (Cucumis About 50% of the farmers sativus L. var. sativus) in this valley. interviewed were over 40 years old and

Table 5. Results of the analyses practiced on selected samples of horticultural crops affected by viral diseases in the Zapotitán Valley, El Salvador.

Sample Crop Analysesa EM MAB-BS MAB-GA PTY1 CMV PCR

1 Pepper + - - + nt nt 2 Pepper + - - + nt nt 3 Tomato - +/- - - nt nt 4 Tomato - ----nt 5 Pepper + - - + nt nt 6 Cucumber + - - + - nt 7 Pepper - + - - nt nt 8 Tomato - - - - nt nt a. EM, electron microscopy; MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses; MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of Bean golden yellow mosaic virus-Guatemala; PTY1, monoclonal antibodies to detect potyviruses; CMV, monoclonal against cucumovirus; PCR, polymerase chain reaction; and nt, not tested.

201 Whiteflies and Whitefly-borne Viruses in the Tropics

40% were between 25-40 years of age. Programa Cooperativo Regional de Only 25% of the farmers interviewed Frijol para Centro América, México y el had secondary (high school) education Caribe [PROFRIJOL] activities in and 41% had received technical El Salvador). A major collaborator in assistance in the selection of this project has been the University of pesticides to control the whitefly El Salvador, under the guidance of problem. Pesticide salesmen are highly Prof. Leopoldo Serrano. The University influential in the selection and use of has generated and collected a insecticides for all crops. Most farmers considerable amount of grey literature (54%) apply insecticides with a on whitefly-related production frequency between 10-15 times per problems in El Salvador and has crop cycle. A significant proportion offered students to conduct thesis (16%) of the farmers interviewed apply research on whitefly-transmitted pesticides more than 20 times in a viruses affecting crops of socio- single crop cycle. One hundred and economic importance in the country. forty-seven different commercial The University of El Salvador covered pesticides were mentioned during this the San Vicente Valley, another survey. Sixty-two percent of the important agricultural area. A second farmers interviewed mentioned institution of higher learning preventive pesticide applications as a collaborating in this project was the risk-averting practice. Universidad Técnica Latinoamericana (UTLA), which also conducted some Of the farmers interviewed, about studies in the Zapotitán Valley. The 97% recognizes the whitefly problem Plant Health Division of the Dirección but only 36% associates this pest with General de Salud Ambiental (DIGESA) viral diseases. Farmers believe that also participated in this collaborative whitefly-related problems are more effort, covering the departments of severe during the dry months of the Santa Ana, Ahuachapán and year but they cannot predict the Sonsonate. Finally, a incidence of whiteflies and viral nongovernmental organization, the diseases based on climatic factors. Programa Regional de Manejo Over 60% of the respondent farmers Integrado de Plagas en América considered that yield losses in Central (PROMIPAC), devoted to the susceptible crops range from production of non-traditional crops, 25%-50% because of the whitefly/ also joined the project by providing begomovirus problems. support to the various activities undertaken by the different groups. The creation of a national network was Strengthened Research unique for Central America and Capacity represented a major local effort to address the important problem of El Salvador was undoubtedly the whitefly-transmitted begomoviruses in country where this project established El Salvador. more collaborative activities with different agricultural research All of these institutions were institutions. The local coordinating provided with technical guidelines for entity was the national agricultural the collection and processing of research institution, CENTA, under biological samples related to the the co-ordination of Dr. Priscila whitefly problem affecting crops of Henríquez (biotechnology) and socio-economic importance. Ing. Carlos A. Pérez (Coordinator of

202 El Salvador

Current Status of The use of virus-resistant common Whitefly/Begomovirus bean cultivars has attenuated the Problems problem in this food crop but genetic resistance is not available in any of the The whitefly/begomovirus problems in other crops affected by begomoviruses El Salvador are widely dispersed in the in El Salvador. main agricultural areas of the country. However, the incidence of these pests is higher in the plains and mid-altitude References valleys. These agro-ecosystems range in altitude, from sea level to 1000 m, with Granillo, C. R.; Díaz, A.; Anaya, M. A.; average temperatures of 27 ºC and Bermúdez, L. A. 1974. Enfermedades transmitidas por about 1600 mm of rainfall; all of these Bemisia tabaci en El Salvador. being conditions that favour the biology Resumen de trabajo, seminario de and reproduction of B. tabaci. leguminosas de grano, Río Piedras, PR. 2 p. The displacement of traditional agriculture (e.g., maize and common Parada, M. E.; Pérez, C. A. 1996. bean) with non-traditional export crops Situación actual del cultivo del frijol común en El Salvador con énfasis (e.g., tomato, pepper and cucurbits) en bacteriosis común. Informe has aggravated the whitefly/ Técnico, Taller de bacteriosis, begomovirus problem because of the Mayagüez, PR. 12 p. availability of new reproductive hosts for B. tabaci and the intensive use of Pastor, R. 1988. Historia de pesticides to protect the new high- Centroamérica. Editorial Piedra Santa, GT. 272 p. value crops against begomoviruses. Zaumeyer, W. J.; Smith, F. F. 1964. The current situation of the Report of a bean disease and insect whitefly/begomovirus problems in El survey in El Salvador. Technical Salvador merits an effort to implement Report, Agency for International integrated pest management strategies Development (AID). Technical in the main horticultural areas of the Assistance Agreement, Agricultural Research Service, United States country, mainly in the departments of Department of Agriculture (USDA), San Vicente, San Salvador, La Beltsville, MD, USA. Libertad, Sonsonate and Santa Ana.

203 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.5 Honduras

Francisco Morales*, María Mercedes Doyle**, Rogerio Trabanino**, Carolina Nolasco**, Elsa Barrientos**, Luís Jara** and Norman Escoto***

Introduction The emergence of Bemisia tabaci as a pest and virus vector The isolated situation of the interior Geographical context valleys and the absence of an extensive The territory of Honduras consists of cotton (Gossypium hirsutum L.) industry highly mineralized soils and mountains, in the Pacific coast of Honduras which favoured the development of probably protected this country from mining and livestock exploitations in early outbreaks of Bemisia tabaci colonial times. Unlike in other Pacific (Gennadius) and begomoviruses Coast valleys of neighbouring countries, recorded in neighbouring countries to there are few volcanoes contributing the north, already in the 1970s. It was their fertile ashes for the benefit of the not until the mid-1980s that the soils. The first region to be devoted to intensification and diversification of agriculture was the humid Caribbean cropping systems in the highland central lowlands or “Costa”, which possess valleys coincided with the emergence of fertile alluvial soils planted primarily to begomovirus problems in Honduras. banana (Musa spp. L.). These Bean golden yellow mosaic virus geographic characteristics leave only the (BGYMV) emerged around 1985. In interior mid-altitude valleys 1989, most common bean (Phaseolus (350-1000 m), for agricultural vulgaris L.) producing regions in production (West and Augelli, 1977). Honduras were already affected by One of the most important valleys in BGYMV. The most affected areas were this mountainous region is the those in the central and southern Comayagua, with an area of about agricultural regions of the country, 295 km2. Much of the mountainous where annual rainfall does not surpass interior of Honduras supports one of the the 1500 mm range, particularly from most extensive pine and oak forests in May through July (Rodríguez et al., Middle America. Figure 1 shows the 1994). main agricultural regions affected by whitefly-transmitted begomoviruses. As in the rest of Middle America, in the last decade Honduras has experienced a boom in the production of non-traditional crops such as tomato * Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. (Lycopersicon esculentum Mill.), pepper ** Escuela Agrícola Panamericana (EAP), (Capsicum spp. L.) and melon (Cucumis Honduras. melo L.). The tomato industry in the *** Departamento de Investigación y Capacitación Agropecuaria (DICTA), Comayagua Valley (580 m) was Honduras. developed as early as 1978, with a view

204 Honduras

Islas de la Bahía Sonora Colón Atlántida Cortés

Honduras Gracias a Dios Santa Yoro Bárbara Copan Comayagua Olancho Ocotepeque Francisco Intibuca Morazán Lempira La Paz El Paraíso

Valle Choluteca

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Honduras.

to producing high-quality produce for scant. During this project, common export (Standard Fruit Company, 1978). bean plants exhibiting golden mosaic- This industry has come practically to an like symptoms were tested with the end now because of the persistent monoclonal BGYMV antibodies. In attacks of whitefly-borne viruses. A these tests, the Honduran BGYMV similar phenomenon has been recorded isolate reacted with the broad for the valleys of Jamastrán, El Paraíso spectrum monoclonal antibody (MAB- and La Lima (El Paraíso Department), BS) but not with the Guatemalan although the most affected crop in these (GA), as observed in this project for areas is pepper. The melon industry of other BGYMV isolates from southern Honduras also has been Guatemala and El Salvador. In a brought to the point of extinction by different survey, eight samples taken whitefly-related problems (Caballero, in the Comayagua Valley from 1995). These production problems have diseased common bean, tomato, forced many growers to abandon the pepper, cucumber (Cucumis sativus L. cultivation of susceptible food crops var. sativus) and a new vegetable leaving as alternative crops such as grown for export, cundeamor tobacco (Nicotiana tabacum L.). However, (Momordica charantia L.), were tobacco is currently under attack by assayed serologically. Table 1 shows whiteflies and begomoviruses in the results obtained with these neighbouring countries, particularly in samples. Guatemala. Begomoviruses were detected in common bean and tomato but not in Advances in Biological the two cucurbits tested, namely Research cucumber and cundeamor. The latter seemed to have a low incidence of a Research on whitefly-transmitted phytoplasma disease, and the begomoviruses in Honduras has been cucumber sample was infected by a

205 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. Results of various diagnostic assays performed on selected plant samples from the Comayagua Valley, Honduras.

Sample Crop Diagnostic assaya EM MAB-BS MAB-GA PTY1 CMV PCR

1 Common bean nt + - nt nt + 2 Cucumber + - nt + nt nt 3 Cundeamor nt - nt nt nt nt 4 Common bean nt + - nt nt + 5 Common bean - - - - - nt 6 Tomato nt + nt nt nt nt 7 Tomato nt + nt nt nt nt 8 Tomato nt + nt nt nt nt a. EM, electron microscopy; MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses; MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of Bean golden yellow mosaic virus-Guatemala; PTY1, monoclonal antibody to detect potyviruses; CMV, monoclonal antibody against cucumoviruses; PCR, polymerase chain reaction; and nt, not tested.

potyvirus, most likely transmitted by of the BGYMV species found in Middle aphids. The BGYMV isolates from America. In order to test the hypothesis Honduras did not react with the MAB- that the differences in nucleotide BS prepared in 1993 to the sequence homology at the coat protein Guatemalan isolate. The original level (AV1) may have occurred in Guatemalan isolate reacted to this response to the presence of the new monoclonal when used as a control in B. tabaci biotype. this test. This result suggests that the present Honduran BGYMV isolates Eight whitefly samples from also have altered their capsid protein common bean, four samples from composition, probably in response to cucumber, three samples from tomato the arrival of a more ubiquitous and one sample from chilli were vector, the B biotype of B. tabaci. biotyped by random amplified polymorphic DNA (RAPD) analysis. The The deoxyribonucleic acid (DNA) results from these tests (Table 3) do not of the BGYMV isolates found infecting support the hypothesis of coat protein common bean was amplified by changes in the Honduran BGYMV polymerase chain reaction (PCR) for isolate due to the emergence of a new cloning and partial sequencing. These whitefly biotype, because only the results (Table 2) show that the A biotype of B. tabaci biotype was Honduran BGYMV isolate is a strain detected in these samples.

Table 2. Partial sequence homology (%) between a Bean golden yellow mosaic virus (BGYMV) isolate from Honduras (HD) and related BGYMV/Bean golden mosaic virus (BGMV) isolates characterized in the early 1990s.

ORFa Isolatesb BGYMV-DR BGYMV-PR BGYMV-GA BGMV-BR BGYMV-HD

AC1 91.1 89.5 92.0 66.8 100 AV1 93.8 91.8 94.2 78.0 100 a. ORF, open reading frame; AC1, viral replicase gene; and AV1, coat protein gene. b. DR, Dominican Republic; PR, Puerto Rico; GA, Guatemala; and BR, Brazil.

206 Honduras

Table 3. Biotyping of whitefly (Bemisia tabaci) samples collected from different commercial crops grown in different municipalities of two departments in Honduras.

Department Municipality Crop Biotype

Comayagua V. San Antonio Common bean A Comayagua V. San Antonio Common bean A Comayagua V. San Antonio Common bean A Comayagua V. San Antonio Common bean A Comayagua V. San Antonio Common bean A Comayagua Comayagua Common bean A Comayagua Comayagua Cucumber A Comayagua Comayagua Cucumber A Comayagua San Nicolás Tomato A Comayagua San Nicolás Tomato A Comayagua Flores Tomato A Comayagua V. San Antonio Common bean A Comayagua Flores Common bean A Comayagua Comayagua Cucumber A Comayagua Comayagua Cucumber A Francisco Morazán Cedros Chilli A

Socio-economic Analysis These results demonstrate that most farmers are aware of the key climatic A case study was conducted in the factors that determine whitefly Comayagua Valley with the epidemics. collaboration of the Escuela Agrícola Panamericana (EAP), Zamorano. The Over 40% of the farmers interviewed results of this study are published in a apply insecticides on a calendar basis thesis (Jara, 1998). In the Comayagua for whitefly control. This practice can Valley, more than 70% of the be interpreted as a risk reduction 100 producers surveyed had been measure, particularly in the case of cultivating the land longer than tomato growers, who invest between 5 years. The number of literate farmers US$2100 and $3500 per manzana in the population surveyed was 75%, of (0.764 ha). In tomato plantings, 63% of which 66% had received only primary the growers apply insecticides against education. Only half or less (in some whiteflies as soon as the tomato plants areas) of the farmers surveyed had are transplanted, 31% apply 1 week after received technical assistance to control transplanting and only 6% wait until whitefly problems. they see whiteflies in their fields. In the case of common bean, 12% of the Regarding whitefly incidence, 87% farmers apply pesticides at planting of the farmers believed that whiteflies time, 42% start controlling whiteflies and whitefly-borne viruses increased in 1 week after germination of the plants the warmer months of the year. About and 46% apply insecticides later on 80% of the farmers noted that whitefly during the vegetative phase of the crop. populations increase in periods of low These results demonstrate farmers’ rainfall. The worst whitefly/virus perception regarding the investment epidemics in the Comayagua Valley required to plant tomato (usually occurred in 1989, when drought and 10 times higher than in the case of high temperatures struck the region. common bean).

207 Whiteflies and Whitefly-borne Viruses in the Tropics

Regarding the likelihood of The Plant Protection Department of introducing non-chemical control the EAP sent Elsa Barrientos to the measures based on cultural practices Centro Internacional de Agricultura such as changing planting dates to Tropical (CIAT), Colombia, for training avoid whitefly population peaks, about in molecular biotyping of B. tabaci and 65% of the growers were reluctant to characterization of plant viruses during change. This finding suggests that a 2-week period. climatic and market factors determine planting times in Honduras. Current Status of In economic terms, 53% of the Whitefly/Begomovirus production costs for tomato and 47% Problems for common bean are related to the chemical control of whitefly/ Although the whitefly/begomovirus begomoviruses. In cucumber problems arrived relatively late in production, only 16% of the total costs Honduras, most of the territory offers are related to chemical whitefly control. suitable environmental conditions for Total production costs for tomato are B. tabaci to reproduce and attack up to 20 times more than for common susceptible crops, either as a pest or as bean or cucumber. It is evident from a virus vector. The exceptions are the this study that the implementation of northern Caribbean coast, which an integrated whitefly management usually receives more than 2000 mm of package in the Comayagua Valley rainfall, a deleterious climatic factor for would significantly contribute to B. tabaci. This region is the prime lessening the environmental impact of banana-producing zone of Honduras. insecticides and to a significant The rest of the country has an average reduction in production costs of high- rainfall of less than 1500 mm and value crops such as tomato and average temperature of 26 ºC, and peppers. most horticultural regions are located between 200 and 1000 m above sea level—all favourable conditions for Strengthened Research B. tabaci. Capacity A good indicator of the presence of The Tropical Whitefly Integrated Pest whitefly-transmitted viruses in a region Management Project made possible the is the occurrence of BGYMV. This virus development of a thesis entitled is distributed in the departments of “Characterization of the incidence and El Paraíso, Francisco Morazán, management of whiteflies Choluteca, Valle, Olancho, Comayagua (Homoptera:Aleyrodidae) in the valley of and Copán, marking a vast region Comayagua, Honduras” at the Plant where whitefly/begomovirus problems Protection Department, EAP-Zamorano. can emerge on susceptible crops. The case study helped acquire a better Undoubtedly, one of the most affected view of how small-scale farmers areas is the Comayagua Valley, north perceive the whitefly problem. Experts of the capital, Tegucigalpa. This valley in various fields such as agronomy, is located in a drier zone (500-1000 entomology, virology, socio-economics mm) and has been planted intensively and biometry developed the to non-traditional export crops under questionnaire, which consequently considerable pesticide abuse. constitutes a guide for future surveys on similar field production problems.

208 Honduras

References Standard Fruit Company. 1978. Reporte final de los cultivos semi- Caballero, R. 1995. Honduras. Reporte, comerciales de tomate y pepino IV Taller Latinoamericano sobre para exportación en los valles de moscas blancas y geminivirus. Comayagua y la Entrada, Copán. Ceiba 36:21-23. La Ceiba, HN.

Jara, L. A. 1998. Caracterización de la West, R. C.; Augelli, J. P. 1977. Middle incidencia y manejo de mosca America: Its lands and peoples. blanca (Homoptera:Aleyrodidae) en Second edition, Prentice-Hall, NJ, el Valle de Comayagua, Honduras. USA. 494 p. Thesis, Departamento de Protección Vegetal, Escuela Agrícola Panamericana (EAP), El Zamorano, HN. 84 p.

Rodríguez, F.; Díaz, O.; Escoto, N. D. 1994. Situación del mosaico dorado en Honduras. In: Morales, F. J. (ed.). Mosaico dorado del frijol: Avances de investigación 1994. Programa Cooperativo Regional de Frijol para Centro América, México y el Caribe (PROFRIJOL)-Swiss Agency for Development and Cooperation (SDC). Centro Internacional de Agricultura Tropical (CIAT), Cali, CO. p. 45-50.

209 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.6 Nicaragua

Francisco Morales*, Alberto Sediles**, Aurelio Llano*** and Falguni Guharayψ

Introduction

Geographical context Boaco and Sébaco where a more Nicaragua is the largest country in intensive agriculture eventually Central America and one of its richest developed, consisting of non-traditional nations in terms of agricultural crops such as tomato (Lycopersicon resources. Its agricultural economy esculentum Mill.), pepper (Capsicum follows the natural divisions of the spp. L.) and other horticultural crops. country: the Pacific region, the Common bean (Phaseolus vulgaris L.) northern part of the Central Highlands remains a major staple and and the lowlands of the Caribbean. The consequently is grown throughout the Pacific region has been Nicaragua’s country. However, the main regions centre of commercial agriculture since producing common bean are the Pacific colonial times, when livestock and the region (30% of total production), the production of indigo (blue dye) were Central Highlands (50%) and the two of the main commodities. Livestock Caribbean Plains (10%). In the production still remains an important highlands, the main departments activity in this region. Cotton producing common bean are Matagalpa, (Gossypium hirsutum L.) became a Jinotega, Estelí, Madriz and Nueva highly important crop in the Pacific Segovia. Figure 1 shows the main lowlands, between Lake Managua and agricultural regions affected by whitefly- the Gulf of Fonseca (Chinandega to transmitted begomoviruses. León) and along the eastern side of Lake Managua (Tipitapa) in the early 1950s. By 1977, Nicaragua was the The emergence of Bemisia largest (217,000 ha) producer of cotton tabaci as a pest and virus in Central America (Gill, 1994). vector Bemisia tabaci (Gennadius) first became As in the rest of Central America, a pest of cotton in the 1970s, which there are a number of fertile mid- together with adverse marketing altitude valleys in Nicaragua such as circumstances reduced the area planted to cotton to a mere 2520 ha in 1993. * Centro Internacional de Agricultura Interestingly, B. tabaci was not an Tropical (CIAT), Cali, Colombia. insect of economic significance in the ** Universidad Nacional Agraria (UNA), early years of cotton production in Managua, Nicaragua. *** Centro Nacional de Investigación Nicaragua. The emergence of this Agropecuaria (CNIA), Nicaragua. whitefly species as a major pest of ψ Centro Agronómico Tropical de cotton followed the introduction and Investigación y Enseñanza (CATIE)-Proyecto Manejo Integrado de Plagas (MIP), El intensive use of pesticides on this crop Zamorano, Nicaragua. in the 1960s (Gill, 1994).

210 Nicaragua

L.) but only bean was affected by begomoviruses. Whitefly-transmitted begomoviruses also attacked pepper and tomato plantings in the Central Nueva Segovia Highlands. Tomato plantings were Madriz Nicaragua affected in the region of Boaco in 1991. Estelí The same year, B. tabaci attacked tobacco (Nicotiana tabacum L.) plantings Chinandega in the departments of Nueva Segovia, León Boaco Jinotega, Estelí, Chinandega, Masaya and Rivas. The melon industry in the departments of León, Rivas, Managua Carazo and Matagalpa was first affected by whiteflies in 1991 but the incidence of viruses was low (Sediles, 1998).

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Nicaragua. Advances in Biological Research

Bean golden yellow mosaic virus No records exist of serological (BGYMV) was first observed affecting identification of BGYMV in Nicaragua common bean in the Pacific Coast of prior to 1995. In 1993, a few common Nicaragua, around 1971. BGYMV is bean and tomato samples were collected particularly severe in Estelí, Nueva near the town of Santa Lucía in the Segovia, León, Chinandega and the Boaco Valley. The common bean lowlands of Matagalpa (Llano et al., samples were taken from plants 1997). Large populations of B. tabaci showing dwarfing symptoms, rather have been reported on cotton in the than mosaic. These samples reacted Pacific Coast of Nicaragua since 1952, positively with the broad spectrum and from tomato in the Atlantic Coast monoclonal antibody that detects as early as 1953 (Hidalgo et al., 1975). whitefly-transmitted begomoviruses in This whitefly species was the third most general (MAB-BS). The virus was later important pest problem in cotton in identified as Bean dwarf mosaic virus Nicaragua in the late 1970s (Kramer, (BDMV) by nucleic acid hybridization 1966). The whitefly problem became so methods (Zamora 1996). At that time, serious that, in 1975, Nicaragua and from the same locality, samples created a special commission to study collected from tomato plants affected by B. tabaci and make recommendations crespo disease also gave positive results on research and whitefly management in serological assays with the same (Hildago et al., 1975). monoclonal antibody. The crespo disease of tomato was first observed in Mid-altitude valleys such as Sébaco Nicaragua in 1986. An in-depth (Matagalpa Department) witnessed the investigation conducted in emergence of whitefly-transmitted 2000 showed that tomato in Nicaragua viruses in 1986. By 1992, yield losses is infected by at least four distinct reached 100% in several tomato fields begomoviruses (Rojas et al., 2000). affected by what growers referred to as “crespo” (leaf curl). In 1991, large In 1995, three BGYMV-affected populations of whiteflies attacked common bean samples from the north- common bean and melon (Cucumis melo western department of Estelí were tested

211 Whiteflies and Whitefly-borne Viruses in the Tropics with the monoclonal antibody used to and Pochocoape) was undertaken to detect the original Middle American collect plant samples for virus assay. isolates of BGYMV-Guatemala (MAB- Table 1 gives the results of this survey, GA), available at the Centro which show a rather low incidence of Internacional de Agricultura Tropical whitefly-transmitted geminiviruses (GV) (CIAT). The three common bean in the samples tested. However, the samples reacted with the MAB-BS but sampling was done at the peak of the only one common bean sample was rainy season, which, for the second recognized by the specific MAB-GA. time since Hurricane Mitch in the previous year, brought considerable A survey in the departments of amounts of rain. Rainfall in 1999 Estelí (Condega), Matagalpa (Apompua almost doubled the amount of rain and Sébaco) and Managua (Managua recorded in 1998 in the Pacific

Table 1. Results of survey of virus-affected plants in three departments of Nicaragua.

Sample Crop Locality Department Reactiona MAB-BS PTY1 1-N5 Tomato Condega Estelí - - 2-N5 Tomato Condega + / - - 3-N5 Tomato Condega - - 4-N5 Tomato Condega + - 5-N5 Tomato Condega - - 6-N5 Tomato Condega - - 7-N6 Tomato Apompua Matagalpa - - 8-N6 Tomato Apompua - - 9-N6 Tomato Apompua - - 10-N6 Tomato Apompua - - 11-N6 Tomato Apompua - - 12-N6 Tomato Apompua - - 13-N7 Tomato Managua Managua - - 14-N7 Tomato Managua - - 15-N7 Tomato Managua - - 16-N7 Tomato Managua - - 17-N8 Tomato Sébaco Matagalpa - - 18-N8 Tomato Sébaco - - 19-N9 Tomato Pochocoape Managua - - 20-N9 Tomato Pochocoape - - 21-N9 Tomato Pochocoape - - 22-N9 Tomato Pochocoape - - 23-N9 Tomato Pochocoape - - 24-N9 Tomato Pochocoape - - 25-N10 Pipián Apompua Matagalpa + + 26-N10 Pipián Apompua + + 27-N10 Pipián Apompua + + 28-N11 Tomato Sébaco Matagalpa - - 29-N11 Tomato Sébaco - - 30-N11 Tomato Sébaco - - 31-N11 Tomato Sébaco - - 32-N11 Tomato Sébaco - - 33-N11 Tomato Sébaco - - 34-N11 Tomato Sébaco - - 35-N12 Tomato Apompua Matagalpa - - 36-N12 Tomato Apompua - - a. Positive or negative reactions to a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses (MAB-BS); reactions to a potyvirus-specific monoclonal antibody (PTY1).

212 Nicaragua

Lowlands of Nicaragua. The presence of traditional food crops from most of the begomoviruses in pipián (Cucurbita fertile valleys in Central America in argyrosperma C. Huber subsp. order to make room for high-value, argyrosperma) is interesting, non-traditional export crops such as considering that this crop had not been tomato, pepper and melon. As a result, recorded previously as an important there is a higher demand for traditional host/reservoir of begomoviruses. crops such as common bean in the However, the same observation was region. Nicaragua could take advantage made in El Salvador and, furthermore, of its relatively larger agricultural area this cucurbit was also doubly infected to capture the increased regional with aphid-transmitted potyviruses in demand for food staples. both countries. Horticultural crops such as tomato Because of the rainy conditions and pepper have been cultivated in that affected Nicaragua in 1998 and Nicaragua mainly to meet local 1999, very few whitefly samples could demand. The area planted to tomato be taken for analyses. Table 2 has doubled in the past 30 years from summarizes some of the preliminary 350 to 750 ha, whereas in Guatemala, results obtained. Although representing tomato production grew from 5000 ha only a limited sample, results are taken in 1960 to 12,000 ha in 1970. In 1986, from one of the main horticultural tomato plantings in the Sébaco Valley, valleys of Nicaragua and, consequently, Matagalpa Department, suffered are interesting in that they show a unusual infestations of the whitefly predominance of the original biotype A. B. tabaci and, soon after, the emergence of viral diseases associated with the whitefly outbreak. By 1991, Table 2. Results of Bemisia tabaci biotyping analyses performed on 15 samples of tomato production in the Sébaco Valley whiteflies from Apompua and had been reduced 20%-50% because of Sébaco, department of Matagalpa, yield losses ranging between 30%- Nicaragua. 100% (Sediles, 1998). A similar Sample Biotype A Other situation was observed for pepper code plantings, which in 1991-92, reported N2 4 1 yield losses between 30%-50%. Melon, another non-traditional crop grown N3 5 - mostly for export, was beginning to N4 5 - experience whitefly-related problems. However, there was no significant incidence of begomoviruses in this crop. Socio-economic Analysis The whitefly problem on tomato Common bean is one of the two main became so severe that it prompted the food staples, together with maize creation of an inter-institutional tomato (Zea mays L.), being grown on about group composed of CATIE/MAG-MIP– 140,000 ha. However, productivity is NORAD-ASDI (for full names, see low (about 450 kg/ha) because 80% of Acronyms list on page 345), which the national production takes place on organized the first national meeting of farms of less than 3 ha. The new tomato producers in April 1994. economic policies that followed the Tomato growers met to discuss whitefly economic crisis of the 1970s have management strategies with the gradually led to the displacement of technical organizing group (Grupo

213 Whiteflies and Whitefly-borne Viruses in the Tropics

Interinstitucional e Interdisciplinario de the hot Pacific Plains, where cotton Tomate-GIIT). In June 1995, a second production took place in past decades. meeting took place in Santa Emilia, The whitefly B. tabaci became a limiting Matagalpa, with 24 producers biotic problem for cotton production and representing 16 tomato-growing regions the region has moved on to new crops: of Nicaragua, where the whitefly is a tomato, watermelon (Citrullus lanatus limiting factor to tomato production. In [Thunb.] Matsum. & Nakai), squash these meetings, tomato farmers (Cucurbita spp. L.), pepper, melon and became aware of the consequences of traditional ones as well such as abusing pesticides as well as of the common bean and tobacco. Soybean existence of integrated pest (Glycine max [L.] Merr.), peanut (Arachis management practices (live barriers, hypogaea L.) and cassava (Manihot yellow traps and organic insecticides). esculenta Crantz) are other crops An outcome of the meeting was the infested by whiteflies in this region. creation of the Nicaraguan Tomato Whitefly population peaks occur in Growers Association, which is January and February. responsible for the sustainability of tomato production in Nicaragua. Region III consists of Managua Department and two vegetable production zones, the coastal area of Strengthened Research Lake Managua located north of the city Capacity of Managua and the zone of Pochocuape, south of Managua. Here, The Universidad Nacional Agraria (UNA) whiteflies attack tomato, squash, located in Managua, the capital of watermelon, common bean and tobacco Nicaragua, was selected as the main and most of these crops are attacked collaborating institution. The national also by begomoviruses. program and the Ministry of Agriculture of Nicaragua, already have Region IV includes the departments the substantial financial support of of Carazo, Granada, Masaya and Rivas, international organizations. The which form the southernmost portion of financial support to UNA permitted a the Pacific Lowlands. At these low general evaluation of the whitefly altitudes, the whitefly B. tabaci thrives, situation in the six administrative severely attacking tomato and tobacco regions of Nicaragua described below. and, to a lesser extent, watermelon, squash, common bean, pepper and Region I includes the north- melon. The main whitefly peak occurs in western departments of Nueva Segovia, January and February. Madriz and Estelí. In this region, there are valleys and plateaus with altitudes Region V is formed by the ranging between 500 and 1000 m and departments of Boaco and Chontales. a variety of food crops such as bean, Although this area has been devoted tomato, cucurbits, tobacco and pepper, primarily to livestock, there are some which attract whiteflies. The whitefly parts such as Santa Lucía, Boaco, population peak occurs in the period where bean and tomato traditionally November-May. have been planted. These crops have been affected by whitefly-transmitted Region II includes the south- begomoviruses since the mid 1980s, western departments of León and particularly by BGYMV (common bean) Chinandega and agricultural lowlands and by the crespo disease in the case of below 500 m. The region is formed by tomato.

214 Nicaragua

Region VI is a main horticultural Caribbean Plains) is only cultivated to a area of Nicaragua, particularly the minor extent with, for example, banana Sébaco Valley, in the department of (Musa spp. L.), rice (Oryza sativa L.) Matagalpa. The second department and oil palm (Elaeis guineensis Jacq.), that forms this region, Jinotega, lies because it is one of the rainiest regions further north, bordering Honduras. The in Central America. This climatic factor tomato crop in this region has been is a major deterrent for the whitefly affected severely by whitefly-borne B. tabaci and for the establishment of begomoviruses, to the point that some its main plant hosts. The Pacific farmers have abandoned this crop, Lowlands, on the other hand, particularly in the southern part of the experience several dry months without Sébaco Valley. Large whitefly significant rainfall and thus possess all populations and viral symptoms also the necessary conditions for the affected cucurbits such as squash, whitefly B. tabaci to thrive, including pumpkin (Cucurbita spp. L.) and the presence of good breeding hosts cucumber (Cucumis sativus L. var. such as cotton. Moreover, as in the rest sativus). of Central America, most horticultural crops are located in mid-altitude It is evident that whiteflies and (200-1000 m) valleys created by the B. tabaci-transmitted viruses affect all mountain ranges that make up the of the agricultural regions of Nicaragua. central and northern regions of Nicaraguan agricultural research Nicaragua. Valleys such as the Sébaco, institutions have come together to Matagalpa Department (450 m) reach affected farmers and manage the constitute an example of a locality with whitefly problem in a highly exemplary a high incidence of B. tabaci and manner. The integrated pest begomoviruses, because of the management projects developed in favourable environmental conditions for Nicaragua with the collaboration of the pest and the presence of susceptible international and national institutions horticultural crops, some of which also have recovered some of the most act as suitable reproductive hosts for affected agricultural areas for the B. tabaci. production of traditional (e.g., common bean) and non-traditional (e.g., tomato) crops. It is expected that the scientific References knowledge generated in these projects will contribute to the implementation of Gill, A. 1994. Problemática del complejo more effective measures to control mosca blanca-virus en algodón en Centroamérica. In: Mata, M.; whiteflies and the viruses they transmit Dardón, D. E.; Salguero, V. E. (eds.). in Nicaragua and Middle America. Biología y manejo del complejo mosca blanca-virosis. Memorias III Taller Centroamericano y del Caribe Current Status of sobre mosca blanca, Antigua, GM. Whitefly/Begomovirus p. 23-38. Problems Hidalgo, O.; León, G.; Lindo, O.; Vaugham, M. 1975. Informe de la Most of the whitefly/begomovirus misión de estudio de la mosca problems in Nicaragua occur in the blanca. Banco Nacional de western half of the country, mainly Nicaragua (BNN), Comisión Nacional because this region concentrates most del Algodón (CONAL), Ministerio de of the country’s population. The Agricultura y Ganadería (MAG), Managua, NI. 120 p. eastern half of Nicaragua (the

215 Whiteflies and Whitefly-borne Viruses in the Tropics

Kramer, P. 1966. Serious increase of Sediles, A. 1998. La mosca blanca cotton whitefly and virus Bemisia tabaci (Gennadius) en transmission in Central America. J. Nicaragua: Historia y problemática. Econ. Entomol. 59:1531. Informe Universidad Nacional Agraria (UNA), NI. 23 p. Llano, A.; Munguía, R.; Viana, A. 1997. Perfil de la producción de frijol en Zamora, M. E. 1996. Identificación de Nicaragua. Instituto Nacional de plantas silvestres como reservorios Tecnología Agropecuária (INTA)- de los virus del mosaico dorado Programa Cooperativo Regional de (BGMV) y del mosaico enano Frijol para Centro América, México (BDMV) del frijol en Pueblo Nuevo, y el Caribe (PROFRIJOL). 33 p. Nicaragua. M.Sc. thesis, (Documento de trabajo) Universidad Nacional Agraria, NI. 84 p. Rojas, A.; Kvarnheden, A.; Valkonen, J. P. T. 2000. Geminiviruses infecting tomato crops in Nicaragua. Plant Dis. 84:843-846.

216 Costa Rica

CHAPTER 3.7 Costa Rica

Francisco Morales*, Luko Hilje**, Juan Vallejos**, Guillermo Sibaja***, Carlos Arayaψ and Rodolfo Arayaψψ

Introduction plantain (Musa × paradisiacal L.). Cotton (Gossypium hirsutum L.) is produced only to a limited extent in the Geographical context Pacific lowlands of the Province of The Central Plateau, around the capital Guanacaste (West and Augelli, 1977; city of San José, is one of the most Pastor, 1988). The Central Plateau is intensively cultivated areas in Costa undoubtedly one of the most affected Rica. North and south of this region, areas in terms of whitefly-related the broken and volcanic topography of damage. Figure 1 shows the main mountain ranges expands all the way agricultural regions affected by to the northern border with Nicaragua whitefly-transmitted begomoviruses. and the southern border with Panama. East and west of the northern mountain ranges lie the Caribbean and Pacific lowlands, the least developed Alajuela and inhabited regions of Costa Rica. Guanacaste Heredia South-eastern Costa Rica receives over 1500 mm of rain annually, which Cartago San José Limón prompted the development of a thriving Puntarenas banana (Musa spp. L.) production Costa Rica industry towards 1880. Coffee (Coffea arabica L.) was introduced into Costa Puntarenas Rica in the 1830s and soon became a source of wealth for many farmers in the Central Plateau. Among the basic food staples, maize (Zea mays L.) and Figure 1. The main agricultural regions affected by whitefly-transmitted common bean (Phaseolus vulgaris L.) begomoviruses, Costa Rica. are important components of the Costa Rican diet, together with crops such as potato (Solanum tuberosum L.) and The emergence of Bemisia tabaci as a pest and virus vector * Centro Internacional de Agricultura The lack of suitable reproductive hosts Tropical (CIAT), Cali, Colombia. ** Centro Agronómico Tropical de for the whitefly Bemisia tabaci Investigación y Enseñanza (CATIE), Costa (Gennadius) such as cotton or soybean Rica. (Glycine max [L.] Merr.), together with *** Ministerio de Agricultura, Costa Rica. ψ Universidad Nacional, Costa Rica. the high rainfall over much of the ψψ Universidad de Costa Rica. Costa Rican territory, protected the

217 Whiteflies and Whitefly-borne Viruses in the Tropics

Central Plateau’s agricultural region Turrialba (Castillo, 1997) and identified against the early arrival of this pest. as Tomato yellow mosaic virus (ToYMV). The first records of the presence of This virus is currently considered a B. tabaci in the country originated in tentative new species referred to as the 1970s, precisely in the drier Tomato yellow mottle virus (ToYMoV). lowlands of the Pacific province of Another begomovirus, isolated from Guanacaste, where an incipient cotton diseased tomato plants in Turrialba, industry was developed in the 1960s. was shown in 1998 to be related to a The first record of B. tabaci as a pest tomato begomovirus tentatively and vector of a plant virus Bean golden designated as Tomato leaf curl Sinaloa yellow mosaic virus (BGYMV) in virus, originally isolated in north- common bean fields in the Central western Mexico (Idris et al., 1999). Valley, occurred in 1987. BGYMV already had been observed in the The Tropical Whitefly Project country in the 1960s (Gámez, 1970), financed a survey of B. tabaci biotypes albeit at a very low incidence. In 1988, in Costa Rica, conducted under the B. tabaci moved on to tomato guidance of Dr. Luko Hilje of the (Lycopersicon esculentum Mill.) in the Centro Agronómico Tropical de Central Valley (Hilje, 1997). Investigación y Enseñanza (CATIE), Coincidentally, the attack to common Turrialba. Table 1 shows the results of bean and tomato took place in the the molecular characterization analyses western areas of the Central Valley, conducted at CIAT for some of the which is the closest region to the samples collected. These results cotton-growing areas of Costa Rica. confirmed the presence of B. tabaci in B. tabaci is currently a pest and vector various crops of socio-economic of viruses in common bean and tomato importance and showed the fields in most of the horticultural penetration of biotype B of B. tabaci in provinces of the country. different regions of Costa Rica. However, at the time of the survey, the original A biotype still predominated in Advances in Biological the country. Some of the most heavily Research colonized crops were cucurbits but, unlike the case of other Central A BGYMV isolate from Alajuela (Central American countries, B. tabaci was Valley) had been characterized found to colonize pepper (Capsicum serologically at the Centro Internacional spp. L.) in Costa Rica (Hilje, 1997; de Agricultura Tropical (CIAT) in 1993 Vallejos, 1997). as a member of the Central American/ Caribbean group of BGYMV isolates. However, different begomoviruses Socio-economic Analysis isolated during the course of this project from BGYMV-affected bean At the beginning of this project, a plants in different regions of Costa Rica previous investigation on the economic (Puriscal, Alajuela and Los Chiles) were impact of the B. tabaci-begomovirus shown to be serologically distinct from pest complex in Costa Rica, and the original Middle American BGYMV particularly on common bean, tomato isolates characterized in the early and pepper, was used as ex ante data 1990s (Cancino et al., 1995). for the survey conducted during the current project (Vallejos, 1997). A begomovirus of tomato had been According to the results obtained in isolated previously in the locality of that investigation, the trend to replace

218 Costa Rica

Table 1. Characterization of Bemisia tabaci biotypes in Costa Rica.

Province Locality Crop Biotype

Alajuela Orotina Watermelon B Alajuela Orotina Cucumber B Puntarenas Puntarenas Melon B Puntarenas Ticaral Watermelon B Puntarenas Ticaral Pumpkin A Puntarenas Lepanto Common bean A/B Guanacaste Carrillo Melon A Guanacaste Carrillo Watermelon A Guanacaste Bagaces Melon A Guanacaste Bagaces Spider flower A Guanacaste Bagaces Cucurbit A Guanacaste Tilarón Tomato A Guanacaste Tilarón Chilli A San José Pérez Zeledón Snap bean A San José Pérez Zeledón Pepper A San José Pérez Zeledón Chilli A San José Pérez Zeledón Tomato A San José Pérez Zeledón Tomato A Heredia not cited Chilli A Heredia not cited Tomato A Heredia not cited Chilli A Heredia not cited Chilli A Limón Guácimo Chilli A Limón Guápiles Pumpkin A Alajuela Gracia Tomato A Alajuela Gracia Chilli A Alajuela Naranjo Chilli A Alajuela Naranjo Cucumber A Alajuela S. Ramón Tomato A Alajuela Atenas Tomato A Alajuela Alajuela Sweetpotato A Alajuela Alajuela Tomato A Alajuela Alajuela Chilli A Alajuela Alajuela Common bean A Alajuela Alajuela Cucumber A Cartago Turrialba Chilli A Cartago Cervante Tomato A Cartago Turrialba Tomato B Cartago Turrialba Chilli A Cartago Cartago Tomato A Cartago Cartago Chilli A Cartago Paraíso Tomato B Cartago Paraíso Chilli B

basic food crops such as common bean expected fall in the productivity of for high-value crops such as tomato these crops. and pepper was quite apparent and still continues in Costa Rica. As a Yield losses due to whitefly- result, common bean and other transmitted viruses were 37% for traditional crops are increasingly common bean, 22% for tomato and 9% planted in marginal soils, with an for pepper. The whitefly B. tabaci could

219 Whiteflies and Whitefly-borne Viruses in the Tropics reproduce on pepper without causing Health Division. Mr. Sibaja came to apparent damage to this host plant. CIAT, Colombia, for intensive training Regarding the economic feasibility of on the characterization of B. tabaci these crops, tomato and chilli biotypes. (Capsicum sp.) were the most profitable crops, producing over US$6300/ha. The net return for common bean was Current Status of US$145/ha. Whitefly/Begomovirus Problems

Strengthened Research Whiteflies and begomoviruses in Costa Capacity Rica mainly affect common bean and tomato crops, particularly in the Costa Rica, and specifically CATIE, has Central Valley, where most of the been a leader in the collection and horticultural crops are grown. Another dissemination of information on area where the whitefly problem is integrated pest management (IPM) gaining momentum is the “Pacífico strategies to control whitefly pests. Seco” (Dry Pacific Region) in Dr. Luko Hilje of CATIE has co- Guanacaste Province. Rainfall in this ordinated these information exchanges region ranges from 1220 to 2000 mm through an international network annually, with an average temperature created in 1992. Annual workshops of 20 to 25 ºC. These are not optimal have been organized in different Latin conditions for B. tabaci but the American countries to update and occurrence of a prolonged dry period discuss whitefly-related problems and from December through April favours the most suitable IPM measures the build up of whitefly populations implemented to date. The Systemwide and transmission of viruses to Tropical Whitefly (TWF)-IPM Project has susceptible horticultural crops. given its support to this effort by Fortunately, most of the Costa Rican promoting complementary activities, territory receives over 1500 mm of particularly in the area of molecular annual rainfall, which maintains characterization of whiteflies and whitefly populations depressed most of begomoviruses. With the funds the year. These climatic factors and provided by this project, Dr. Hilje was various IPM measures implemented in able to attend the International Costa Rica, including the use of Workshop on Bemisia and resistant varieties in the case of Geminiviruses held in San Juan, common bean, have contributed Puerto Rico, in June 1998. His significantly to the attenuation of the presence in this meeting was important whitefly/begomovirus problems in this for maintaining the Latin American country. whitefly network active and for planning future activities. Additionally, the funds provided to CATIE allowed References Dr. Hilje to conduct a countrywide survey of the whitefly/begomovirus- Cancino, M.; Abouzid, A. M; Morales, F. J; affected regions in Costa Rica. Purcifull, D. E; Polston, J. E.; Hiebert, E. 1995. Generation and characterization of three A short study fellowship was monoclonal antibodies useful in awarded to Ing. Agr. Guillermo Sibaja detecting and distinguishing Bean Chinchilla of the Costa Rican Ministry golden mosaic virus isolates. of Agriculture and Livestock, Plant Phytopathology 85:484-490.

220 Costa Rica

Castillo, J. 1997. Movimiento diario de Pastor, R. 1988. Historia de Bemisia tabaci en parcelas de Centroamérica. Editorial Piedra tomate, diseminación local del Santa, GT. 272 p. mosaico amarillo y fuentes de inoculo del ToYMV-CR en Guayabo, Vallejos, J. E. 1997. Sistema experto para Costa Rica. M.Sc. thesis, Centro la evaluación del impacto del Agronómico Tropical de complejo Bemisia tabaci- Investigación y Enseñanza (CATIE), geminivirus, en frijol, tomate y chile Turrialba, CR. 93 p. dulce con fines de planificación. M.Sc. Thesis, Centro Agronómico Gámez, R. 1970. El virus del moteado Tropical de Investigación y amarillo del frijol, plantas Enseñanza (CATIE), Turrialba, CR. hospederas y efecto en producción. 109 p. In: XV Reunión Annual PCCMCA, Antigua Guatemala, GT. p. 44-48. West, R. C.; Augelli, J. P. 1977. Middle America: Its lands and peoples. Hilje, L. 1997. Posibilidades para el Second edition. Prentice-Hall, NJ, manejo integrado del complejo USA. 494 p. Bemisia tabaci-geminivirus en Costa Rica. Agron. Costarricense 21: 139-142.

Idris, A. M.; Rivas-Platero, G.; Torres- Jerez, I.; Brown, J. K. 1999. First report of Sinaloa tomato leaf curl geminivirus in Costa Rica. Plant Dis. 83:303.

221 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.8 Panama

Francisco Morales*, Orencio Fernández**† and José Guerra**

Introduction shows the main agricultural regions affected by whitefly-transmitted begomoviruses. Geographical context Most of the Panamanian territory The emergence of Bemisia receives tropical rains in excess of tabaci as a pest and virus 2500 mm/yr, a climatic factor known vector to drastically reduce whitefly The presence of Bemisia tabaci populations. Nevertheless, during the (Gennadius) on tomato (Lycopersicon first (1992) workshop on whiteflies in esculentum Mill.) was first observed in Central America and the Caribbean, the Azuero Peninsula in 1983. held in Costa Rica, Panama was However, it was not until 1991, when already present (Zachrisson and unusually dry climatic conditions Poveda, 1992). According to the report favoured the reproduction of B. tabaci, presented by the Panamanian delegate that this whitefly species became a pest at that meeting, the emergence of in the central provinces of Panama. whiteflies as agricultural pests in B. tabaci has been observed to Panama was associated with the reproduce in eggplant (Solanum expansion of non-traditional crops, melongena L.), sweet pepper (Capsicum mainly melon (Cucumis melo L.), for annuum L. var. annuum) and chilli export to the United States. Three pepper (Capsicum spp. L.). In 1992, the horticultural areas were developed in silver leaf symptom was observed on the late 1980s: the Pacific Coast of the squash (Cucurbita spp. L.) attacked by province of Panama; the central B. tabaci, which suggested the presence provinces of Coclé, Herrera and Los of biotype B in Panama. The western Santos (the largest); and the western horticultural zone, located in Chiriquí province of Chiriquí, near the Costa Province, is at a higher elevation and Rican border (Zachrisson and Poveda, receives more rainfall than the other 1992; Delgado, 1994). These regions two horticultural zones of Panama. The have in common the presence of a well- problem pest in this area is not so defined dry period and below-average much B. tabaci as Trialeurodes precipitation during the year. Figure 1 vaporariorum (Westwood), a whitefly species that attacks horticultural crops at altitudes above 1000 m (Zachrisson * Centro Internacional de Agricultura Tropical and Poveda, 1992). (CIAT), Cali, Colombia. ** Instituto de Investigación Agropecuaria de Panamá (IDIAP). In Panama, common bean † R.I.P. (Phaseolus vulgaris L.) is cultivated in

222 Panama

Colón

Comarca de Panama San Blas Colón Bocas del Toro Panama

Chiriquí Coclé

Veraguas

Darién Herrera Panama Los Santos

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Panama.

Chiriquí Province, at altitudes between horticultural zone of Azuero was 550 and 1000 m. Bean golden yellow conducted at the Centro Internacional mosaic virus (BGYMV) was observed de Agricultura Tropical (CIAT), with affecting common bean in Panama samples provided by Ing. Agr. José prior to 1970 (Gámez, 1970) but this Angel Guerra of the Instituto de disease never has been a major Investigación Agropecuaria de Panamá production problem of common bean in (IDIAP) (Table 1). Results demonstrate Panama. the coexistence of biotypes A and B of B. tabaci in the central provinces of the Azuero Peninsula. This is an Advances in Biological interesting situation that needs to be Research monitored in order to study the outcome of the interaction between A whitefly-borne geminivirus affecting these two biotypes in a horticultural tomato was described in Panama in zone characterized by mixed cropping 1998 (Engel et al., 1998). The virus, systems. named Tomato leaf curl virus, was shown in this project to have a high degree of nucleotide sequence Socio-economic Analysis similarity when compared to Tomato yellow mosaic virus (ToYMV) from Because Panama was not originally Venezuela. The virus had been included in this project, no socio- described in Panama as a strain of economic information was collected for Potato yellow mosaic virus, a misnomer this country. However, it is known that for ToYMV. Although this virus may be whitefly-transmitted geminiviruses do a new species, the name “Tomato leaf not affect common bean severely in curl virus” is not appropriate because it Chiriquí Province. is already given to a distinct begomovirus species that only exists in Between 1991 and 1994, about the Old World (Asia). Therefore, the 6000 tons of tomato were lost to Panamanian virus should be renamed. whitefly problems in Panama. During the 1997-98 tomato planting season, A molecular characterization of 2800 tons were lost because of the B. tabaci biotypes from the central effect of B. tabaci, both as a pest and

223 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. Molecular characterization of Bemisia These scientists were trained in the tabaci biotypes from Azuero, Panama. molecular characterization of B. tabaci Sample Crop Biotype biotypes and the use of monoclonal 1 Tomato A/B antibodies for the identification of 2 Melon A begomoviruses. 3 Melon B 4 Melon A/B 5 Melon A Current Status of 6 Tomato B Whitefly/Begomovirus 7 Tomato B Problems 8 Tomato B 9 Tomato B Although Panama is one of the rainiest 10 Tomato A/B countries in the world, with an average 11 Melon A/B annual rainfall above 2000 mm, there 12 Melon B are less humid (1000-2000 mm) 13 Melon B coastal regions around the Gulf of Panama, where horticultural 14 Melon B production has increased significantly 15 Melon B in the last decades. The main 16 Melon B horticultural area is located in the 17 Melon B Azuero Peninsula, where most of the 18 Melon A/B whitefly/begomovirus problems have emerged in past years. Whitefly populations increase in the drier vector of plant viruses. Whereas large- months of the year: January-March. scale tomato growers absorbed most of Panama is in the process of expanding the losses, these pests affected a the production of non-traditional number of small-scale producers in the export crops and consequently this province of Los Santos. country should take measures to control the whitefly problem before it becomes unmanageable. Strengthened Research Capacity References Considering that Panama was not included officially in the project, an Delgado, A. 1994. Cultivos hortícolas en agreement was made with the Red Panamá: Algunos datos Colaborativa de Investigación y estadísticos. Folleto informativo, Instituto de Investigación Desarrollo de Hortalizas para América Agropecuaria de Panamá (IDIAP), Central, Panamá y República PA. 17 p. Dominicana (REDCAHOR), a horticultural research network Engel, M.; Fernández, O.; Jeske, H.; operating in the region, to train Frischmuth, T. 1998. Molecular Panamanian researchers. The network characterization of a new whitefly- financed the training at CIAT of two transmissible bipartite geminivirus infecting tomato in Panama. J. Gen. national scientists: an entomologist Virol. 79:2313-2317. (Ing. José Guerra) and a virologist (Dr. Orencio Fernández) from IDIAP.

224 Panama

Gámez, R. 1970. Los virus del frijol en Zachrisson, B.; Poveda, J. 1992. Las Centroamérica. I. Transmisión por moscas blancas en Panama. In: moscas blancas (Bemisia tabaci Hilje, L.; Arboleda, O. (eds.). Las Gen.) y plantas hospedantes del moscas blancas (Homoptera: virus del mosaico dorado. Turrialba Aleyrodidae) en América Central y el 21:22-27. Caribe. Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Turrialba, CR. p. 64-66.

225 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.9 Haiti

Francisco Morales*, Jackson Donis** and Emmanuel Prophete**

Introduction crops have become increasingly important in Haiti, mainly as an extension of the agricultural export Geographical context business in the Dominican Republic. Haiti is a mountainous country; 40% of Figure 1 shows the main agricultural its territory is located at altitudes above regions affected by whitefly-transmitted 490 m. The mountain ranges alternate begomoviruses. with fertile valleys, the largest being the Plaine du Nord. Annual rainfall varies The emergence of Bemisia from 500 in the north-west to tabaci as a pest and virus 2500 mm in the eastern/southern vector highlands. Agriculture accounts for The whitefly Bemisia tabaci was first one-third of the gross domestic product regarded as a serious pest in Haiti, in but subsistence farming predominates. the early 1980s. Common bean, tomato Maize (Zea mays L.; 250,000 ha), (Lycopersicon esculentum Mill.), sorghum (Sorghum bicolor [L.] Moench; eggplant (Solanum melongena L.), 160,000 ha), common bean (Phaseolus pepper (Capsicum spp. L.), cowpea vulgaris L.; 100,000 ha) are the main (Vigna unguiculata [L.] Walp.) and lima staples, with rice (Oryza sativa L.) and bean (Phaseolus lunatus L. var. sweetpotato (Ipomoea batatas [L.] Lam.) lunatus) are the main crops affected by also being important components of this insect pest (Donis and Prophete, the Haitian diet. However, most of the 1997). As a vector, B. tabaci already food consumed in Haiti has to be had been observed to transmit Bean imported. Coffee (Coffea arabica L.) is golden yellow mosaic virus (BGYMV) in the main export crop (145,000 ha), 1978 (Balthazar, 1978). followed by sugarcane (Saccharum officinarum L.), sisal (Agave sisalana Perrine ex Engelm.) and cacao Advances in Biological (Theobroma cacao L.). French-Creole is Research spoken by 90% of the population and only 10% speak French. Horticultural The main agricultural areas of Haiti were surveyed and plant samples analysed to determine the importance * Centro Internacional de Agricultura Tropical of whiteflies and whitefly-borne viruses. (CIAT), Cali, Colombia. Table 1 shows the different regions ** Centre de recherche et de documentation visited and the results obtained from agricoles-Ministère de l’agriculture des ressources naturelles et du développement the tests practiced with samples rural (CRDA-MARNDR), Haiti. collected. Of whitefly-transmitted

226 Haiti

Nord-Ouest

Nord Nord-Est

Haiti Artibonite Centre

Ouest

Grande-Anse Ouest Grande-Anse Ouest

Sud Sud-Est

Sud

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Haiti. begomoviruses, BGYMV was present in sequenced and compared to the rest of common bean and Tomato yellow leaf BGYMV isolates from the Caribbean curl virus (TYLCV) in tomato. As in the and Central America characterized to case of the Dominican Republic, the date. Table 2 shows the results of these specific monoclonal antibody (MAB- comparative analyses. GA), which recognizes BGYMV isolates from Central America and southern As suggested by the serological Mexico, detected the Haitian isolate of assay using the Guatemalan (GA) BGYMV. This isolate was partially specific MAB, the Haitian BGYMV

Table 1. Results of serological and polymerase chain reaction (PCR) assays of selected samples collected in Haiti.

Sample Locality Plant Reactiona MAB-BS MAB-GA PTY1 TYLCV

1 Cul-de-Sac Cowpea - nt - nt 2 Passereine Tomato - nt - + 3 Passereine Tomato - nt - + 4 Passereine Tomato - nt - + 5 Passereine Cowpea - nt + nt 6 Passereine Common bean + + - nt 7 Terrier Rouge Tomato - nt + + 8 Terrier Rouge Tomato - nt - + 9 St Raphael Tomato - nt - + 10 St Raphael Tomato - nt - + 11 St Raphael Pepper - nt + nt a. Positive or negative reactions to a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses (MAB-BS); a MAB used to detect the original Middle American isolates of Bean golden yellow mosaic virus-Guatemala (MAB-GA); a potyvirus-specific monoclonal antibody (PTY1); and PCR detection of Tomato yellow leaf curl virus; and nt, no test.

227 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 2. Comparative nucleotide sequence homology (%) between a begomovirus isolated from common bean in Haiti and previously sequenced common bean begomoviruses.

Region BGa Haiti Common bean begomovirusesb BGYMV BGYMV BGYMV BGMV BDMV Dominican Guatemala Puerto Rico Brazil Colombia Republic

Coat protein (AV1) 100 90.2 92.0 89.5 68.3 68.7

Replicase (AC1) 100 87.5 90.9 88.9 77.6 78.8 a. BG, begomovirus. b BGYMV, Bean golden yellow mosaic virus; BGMV, Bean golden mosaic virus; and BDMV, Bean dwarf mosaic virus. isolate assayed is similar to the Central valleys below this altitude, where American and Caribbean BGYMV B. tabaci thrives, cultural practices isolates but different from the South such as the implementation of a period American bean begomoviruses, Bean free of certain crops that act as golden mosaic virus and Bean dwarf reproductive hosts for the whitefly mosaic virus. Few of the tomato, pepper should be adopted following the model and cowpea samples collected in Haiti implemented in the Dominican contained aphid-transmitted Republic. Although there are tomato potyviruses. All tomato samples were varieties resistant to TYLCV, these are shown by polymerase chain reaction commercial varieties that small-scale (PCR) to be infected by TYLCV. The farmers cannot afford. The production composite whitefly sample collected of high-value crops (e.g., tomato, from eggplant resulted in the detection pepper and eggplant) by small-scale and identification of biotype B of farmers could greatly benefit from the B. tabaci. Although preliminary, these adoption of integrated pest results suggest that the whitefly and management (IPM) practices such as begomovirus problems in the the use of micro-tunnels and biological Hispaniola Island (Dominican Republic insecticides or mild soaps. and Haiti) are similar. Strengthened Research Socio-economic Analysis Capacity

Whitefly-transmitted begomoviruses Two collaborating scientists from the have been observed to cause significant Centre de recherche et de or even total yield losses in common documentation agricoles-Ministère de bean and susceptible horticultural l’agriculture des ressources naturelles crops in Haiti. In the case of common et du développement rural (CRDA- bean, the cultivation of BGYMV- MARNDR, Damien), participated in the resistant varieties (available through the field survey conducted in Haiti. These Programa Cooperativo Regional de Frijol scientists were invited to attend para Centro América, México y el Caribe international workshops in Central [PROFRIJOL] project) and the America and to acquire new whitefly/ cultivation of common bean at altitudes begomovirus characterization above 1000 m should reduce the techniques at the Centro Internacional incidence of BGYMV. The same strategy de Agricultura Tropical (CIAT) but could be adopted for begomovirus- official and administrative problems susceptible horticultural crops, that is, prevented their participation in these growing vegetables above 1000 m. For activities. Nevertheless, as

228 Haiti collaborators in this project, these low economic capacity of most small- researchers have benefited from the scale farmers practicing subsistence information generated to date. agriculture also precludes the use of effective but expensive insecticides The visit to Haiti was also helpful (e.g., imidacloprid) to control whiteflies for establishing closer links between and whitefly-transmitted viruses. The the PROFRIJOL project, CIAT and low levels of insecticidal action of Haiti. As a result, two bean breeders cheaper pesticides encourages are now actively working to deploy pesticide abuse, with the consequent BGYMV-resistant bean varieties in development of insecticide resistance in Haiti, and the United States Agency for B. tabaci populations. Finally, Haiti International Development (USAID) has might benefit from its broken financed a new project to work on topography to avoid the attack of sustainable agricultural practices, B. tabaci by planting susceptible crops including hillside technologies and food at altitudes above 1000 m. security issues in Haiti. References Current Status of Whitefly/Begomovirus Balthazar, S. 1978. Les viroses du haricot commun en Haiti. Fondo Problems Salvadoreño de Estudios de l’Agriculture, des Ressources The importance of B. tabaci as a pest Naturelles et du Développement and vector of plant viruses is expected Rural (DARNDR)-Servicio Nacional to decrease in the near future as new de Sanidad Agropecuária (SENASA), bean cultivars possessing resistance to Damien, SV. BGYMV are introduced by the USAID- Donis, J.; Prophete, E. 1997. Las moscas funded project. The introduction of blancas (Homoptera:Aleyrodidae) en TYLCV-resistant tomato varieties from Haití: Situación actual y manejo. the Dominican Republic should also In: Memoria del VI Taller reduce the impact of TYLCV in Haiti. Latinoamericano y del Caribe sobre However, poverty issues will limit the moscas blancas y geminivirus, adoption of resistant cultivars, 18-19 agosto 1997, Santo Domingo, particularly in the case of tomato. The DM.

229 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 3.10 Cuba

Francisco Morales*, Gloria González**, Carlos Murguido**, Ana Echemendía**, Yolanda Martínez***, Yenín Hernández***, Benito Faureψ and María Chaillouxψ

Introduction var. sativus), cabbage (Brassica oleraceae L.), sweetpotato (Ipomoea batatas [L.] Lam.), eggplant (Solanum Geographical context melongena L.) and common bean About 25% of Cuba’s territory is (Phaseolus vulgaris L.). Vegetable mountainous, with three distinct production has become increasingly mountain ranges running east-west: important to support a rapidly the eastern (Sierra Maestra), central expanding tourist industry in Cuba. (alturas) and western (Cordillera de Guaniguanico) ranges. The remaining The emergence of Bemisia land is composed of extensive plains tabaci as a pest and virus and basins. Cuba has a semi-tropical vector climate with two seasons: dry from The emergence of Bemisia tabaci November through April and rainy from (Gennadius) as a pest is probably May through October. The mean linked to the advent and intensive use annual temperature is 26 °C, with a of agricultural pesticides soon after 23-28 °C range. The average World War II. B. tabaci was reported as precipitation is 1380 mm. All these a pest of tobacco (Nicotiana tabacum L.) conditions favour the dissemination of and a vector of plant viruses of whiteflies, particularly during the dry common bean in the mid 1970s season of the year, particularly in the (Blanco and Bencomo, 1978). But it regions marked in Figure 1. was not until 1989 that B. tabaci became a major production problem of Horticultural products have been tomato and common bean, as a vector traditional commodities in Cuba, of begomoviruses (Murguido et al., particularly those referred to as 1997). Currently, this whitefly species viandas and basic grains. These crops attacks tomato, common bean, squash, include tomato (Lycopersicon cucumber, melon (Cucumis melo L.), esculentum Mill.), squash (Cucurbita cabbage and eggplant throughout spp. L.), cucumber (Cucumis sativus L. Cuba.

By 1990, the new B biotype of * Centro Internacional de Agricultura B. tabaci already had been introduced Tropical (CIAT), Cali, Colombia. ** Instituto de Investigaciones de Sanidad into Cuba and surrounding Caribbean Vegetal (INISAV), Cuba. islands (Brown, 1994). Between 1991 *** Centro Nacional de Sanidad Agropecuaria and 1993, the silver leaf syndrome (CENSA), Cuba. ψ Instituto de Investigaciones de Hortícolas induced by this new biotype was “Liliana Dimitroya” (LILIANA), Cuba. observed on squash (Murguido et al.,

230 Cuba

Ciudad de la Habana

La Habana Matanzas Villa Clara Pinar del Río Cienfuegos Ciego de Sancti Avila Spiritus Las Tunas Camagüey Isla de la Holguín Juventud Cuba

Granma Santiago de Cuba Guantánamo

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Cuba.

1997). Another major production Province were also shown to consist of problem for Cuba and the Caribbean B. tabaci biotype B. These findings region occurred when and Old World suggest that biotype B has adapted virus, Tomato yellow leaf curl virus well and now predominates over (TYLCV) was introduced from Israel biotype A of B. tabaci in Havana into the Caribbean region, including Province. Cuba (Polston and Anderson, 1997). This exotic virus has caused crop Bean golden yellow mosaic virus losses worth millions of dollars in the (BGYMV) has been an important Caribbean region, and it is now known disease of common bean in Cuba since to occur in southern USA and the the early 1970s (Blanco and Bencomo, Peninsula of Yucatán, Mexico. 1978). The epicentre of the problem was the locality of Velasco in the western province of Holguín and it is Advances in Biological now widely disseminated in the Research provinces of Holguín, Las Tunas, Guantánamo and Havana. In recent A main objective of the research years, the incidence of BGYMV has undertaken in Cuba was to determine increased in the provinces of Ciego de the composition of the whitefly Avila, Holguín, Las Tunas and population. A Cuban trainee, Ms. Yenín Camagüey. Hernandez of the Centro Nacional de Sanidad Agropecuaria (CENSA) When the monoclonal antibodies conducted the tests at the Centro (MABs) to BGYMV were developed in Internacional de Agricultura Tropical 1990 (Cancino et al., 1995), the Cuban (CIAT) in Colombia, under the isolate of BGYMV reacted with a supervision of CIAT technical staff. monoclonal antibody (MAB 2G5) Ninety-nine samples were collected produced to a Guatemalan isolate of from tomato plants in the localities of BGYMV, which recognized all of the Quivicán and Alquizar, Havana Middle American BGYMV isolates. By Province. The results of the random 1993, the Cuban isolates of BGYMV amplified polymorphic DNA (RAPD) were not reacting with the specific MAB assay showed that 100% of the whitefly 2G5. As can be observed in Table 1, samples tested corresponded to biotype none of the BGYMV samples from three B of B. tabaci. Additionally, seven different provinces of Cuba reacted whitefly samples from potato (Solanum with the specific MAB-GA (2G5), tuberosum L.) plants grown in Havana including a BGYMV isolate from lima

231 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. Assay of selected common bean samples with a broad spectrum and specific monoclonal antibodies prepared to Bean golden yellow mosaic virus (BGYMV), Guatemala, 1993.

Sample Plant Locality MAB-BSa MAB-GAb

1 Common bean Velasco, Holguín + - 2 Lima bean Velasco, Holguín + - 3 Common bean (Bonita 11) Tomeguín, Matanzas + - 4 Common bean (Chévere) Tomeguín, Matanzas + - 5 Common bean (Velasco Largo) Tomeguín, Matanzas + - 6 Common bean (BAT 304) Pulido, La Habana + - 7 Common bean CIATc ++ 8 Common bean CIATd -- a. MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses. b. MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of BGYMV- Guatemala. c. Sample 7, BGYMV-GA, Guatemala; CIAT, Centro Internacional de Agricultura Tropical. d. Sample 8, healthy common bean. bean (Phaseolus lunatus L.). The whereas the Dominican BGYMV homologous control, BGYMV-GA, isolates had not. In order to investigate reacted with MAB 2G5 as expected. whether the failure of the MAB-GA to However, all of the diseased plant detect the Cuban BGYMV isolates was samples that were assayed reacted due to a major change in the capsid with the broad spectrum MAB (3F7) to protein or putative replicase virus BGYMV, which recognizes bipartite genes, one of the BGYMV isolates from begomoviruses. Quivicán was partially sequenced. Table 3 shows the result of this These tests were repeated in 1997, analysis. These results suggest that the including some BGYMV-infected Cuban BGYMV isolate has not changed common bean plants from the significantly in relation to the original Dominican Republic, a neighbouring BGYMV isolates from the region and Caribbean island (Table 2). The results that the Cuban isolate still can be show that the Cuban BGYMV isolates considered a strain of BGYMV. had changed their antigenic properties,

Table 2. Assay of selected common bean samples with a broad spectrum and specific monoclonal antibodies prepared to Bean golden yellow mosaic virus (BGYMV), Guatemala, 1997.

Sample Plant Locality MAB-BSa MAB-GAb

1 Common bean Quivicán + - 2 Common bean Quivicán + - 3 Common bean Quivicán + - 4 Common bean Dominican Republic + + 5 Common bean Dominican Republic + + 6 Common bean CIATc ++ 7 Common bean CIATd -- a MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses. b. MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of BGYMV- Guatemala. c. Sample 6, BGYMV-GA, Guatemala; CIAT, Centro Internacional de Agricultura Tropical. d. Sample 7, healthy common bean.

232 Cuba

Table 3. Comparative homologies (%) between a Cuban Bean golden yellow mosaic virus (BGYMV) isolate and other common bean begomoviruses.

ORFa BGYMV BGMV-BRb BDMV-COc Guatemala Dominican Republic Puerto Rico

AC1 91.5 91.5 89.8 71.9 69.9 AV1 91.3 91.6 91.3 81.6 78.1 a. ORF, open reading frame; AC1, replicase; and AV1, capsid protein. b. BGMV-BR, Bean golden mosaic virus-Brazil. c. BDMV-CO, Bean dwarf mosaic virus-Colombia.

With respect to tomato, one of the In 1998, Dr. Gloria González of the most affected crops in Cuba, three Instituto de Investigaciones de Sanidad random samples were take in the Vegetal (INISAV), Havana, Cuba, locality of Quivicán, Havana Province, collected samples from symptomatic and tested for different pathogens potato plants in Havana Province. As (Table 4). As shown, the broad reported above, potato in Cuba is spectrum monoclonal antibody (MAB- colonized also by the B biotype of BS) used to detect bi-partite B. tabaci. For this test, the viral nucleic begomoviruses did not detect any acid extracted from affected potato bipartite begomovirus in the tomato plants was amplified by PCR using the samples, although the observation of primers developed by Rojas (1992). The two of the samples in the electron amplified region (AV1 and AC1) was microscope revealed the presence of cloned and sequenced. Table 5 shows begomovirus-like particles. This result the results of the comparative suggested the need to test for the nucleotide sequence analyses presence of TYLCV, a monopartite conducted in reference to other begomovirus reported to attack tomato whitefly-transmitted begomoviruses. in Cuba (Polston and Anderson, 1997). This introduced virus is best detected As can be observed from this by polymerase chain reaction (PCR), comparative test, the begomovirus using specific primers, kindly provided isolated from potato in Cuba is an by D. P. Maxwell and M. K. Nakhla, of isolate of Tomato mottle Taino virus the Department of Plant Pathology, (ToMoTV), a virus described in 1997 University of Wisconsin. The PCR assay attacking tomato in Cuba (Ramos et detected the presence of TYLCV in two al., 1997). This virus was also closely of the tomato samples from Quivican, related to Bean dwarf mosaic virus from Cuba. Colombia (93.3%), Tomato yellow

Table 4. Diagnostic assays practiced with three diseased tomato samples from Quivicán, Havana, Cuba.

Sample EMa MAB-BSb PTY1c CMVd PCR-TYLCVe

1 Isometric - - - + 2 Isometric - - - + 3- - -- - a. EM, electron microscopy. b. MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses. c. PTY1, potyviruses. d. CMV, cucumoviruses. e. PCR-TYLCV, DNA amplification of Tomato yellow leaf curl virus.

233 Whiteflies and Whitefly-borne Viruses in the Tropics mosaic virus (ex-Potato yellow mosaic Tabaquina, a concoction of tobacco virus) from Venezuela, Tomato mottle leaf residues from the intensive Cuban virus from Florida, USA, Sida golden tobacco industry, is another industrial mosaic virus from Costa Rica and subproduct used in Cuba to control Abutilon mosaic virus (South American whiteflies in a highly effective manner. isolate and type species of the cluster Obviously, the well-planned cropping in which the Cuban tomato virus is systems in Cuba also facilitate the taxonomically placed). implementation of legal measures regulating the time of planting for

Table 5. Comparative nucleotide sequence certain crops which act as reproductive homology (%) between a begomovirus host for the whitefly B. tabaci isolated from potato in Cuba and (Murguido et al., 1997). previously sequenced whitefly- transmitted begomoviruses. The control of BGYMV in beans, on a Region Virus the other hand, is effectively carried ToMoTV ToYMV SiGMV out through the use of BGYMV- Coat protein (AV1) 98.5 84.2 85.1 resistant bean cultivars developed in Replicase (AC1) 97.7 83.7 82.4 collaboration with the Programa a. ToMoTV, Tomato mottle Taino virus; ToYMV, Cooperativo Regional de Frijol para Tomato yellow mosaic virus; and SiGMV, Centro América, México y el Caribe Sida golden mosaic virus. (PROFRIJOL) network and CIAT- Colombia.

Socio-economic Analysis Strengthened Research Cuba has a unique comparative Capacity advantage over the rest of the Latin American countries affected by The two main Cuban institutions whiteflies and whitefly-borne viruses, collaborating in the Tropical Whitefly specifically, the limited use of agro- Integrated Pest Management (TWF-IPM) chemicals in this island. Moreover, Project were INISAV and the Instituto Cuba has been able to develop and de Investigaciones de Hortícolas successfully implement biological “Liliana Dimitroya” (LILIANA). During control methods to combat the whitefly Phase I, a third Cuban institute, B. tabaci. One of the most effective CENSA, joined the project. The entomopathogens produced has been sub-project’s coordinator for Cuba, Verticillium lecanii. This fungus is very Dr. Gloria González of INISAV, received pathogenic to the immature stages of funds to attend the VII Latin American B. tabaci. The production of biological and Caribbean Workshop on whiteflies control agents in Cuba constitutes an and begomoviruses held on industrial activity in at least 26-30 October 1998, in Nicaragua. 15 provinces of Cuba, where it is being Dr. González also visited CIAT, applied on a regular basis to Colombia, for a short but intensive horticultural crops affected by training period on the characterization B. tabaci. In 1996, 170 tons of the of whitefly-transmitted viruses. fungus compound were produced in Cuba to treat 12,565 ha of affected The TWF-IPM Project contributed horticultural crops. During the peak of to the organization of an International the 1992 whitefly epidemics, 29,896 ha Workshop on Begomoviruses in the were treated in Cuba with this bio- Caribbean Region, held in Quivicán, control agent (Murguido et al., 1997). Cuba on 24-29 November 1997. The

234 Cuba project provided three keynote Granma, Camagüey, Villa Clara, speakers: the coordinator, Dr. Cienfuegos and Sancti Spiritus. Francisco J. Morales; the past coordinator of the TWF-IPM Project, The provinces of Holguín and Dr. Pamela K. Anderson; and the main Las Tunas are located in one of the support scientist for the Caribbean regions with less precipitation. Also, region, Dr. Jane Polston of the these provinces have high average University of Florida. Dr. Anderson temperatures (23.5 ºC) during the explained the nature and objectives of winter season (January), conditions the TWF-IPM Project; Dr. Polston that favour the reproduction of B. delivered a talk on whitefly-transmitted tabaci. The eastern end of the island viruses in the Caribbean Region and concentrates the bulk of the tobacco disserted on advanced techniques for and vegetable production of Cuba, the characterization of begomoviruses; which probably favours the and Dr. Morales gave a talk on development of large whitefly breeding for disease resistance to populations. whitefly-transmitted viruses. The use of entomopathogens, The Whitefly Project also provided mainly the fungus Verticillium lecanii, funds for the training at CIAT-Colombia constitutes a highly effective control of a junior scientist, Ms. Yenín measure against the proliferation of Hernández, of CENSA, Cuba, on the B. tabaci. Unfortunately, the molecular characterization of B. tabaci production of this micro-organism does biotypes. CENSA is one of the leading not meet the needs of the intensive institutes in Cuba in the area of agriculture practiced in Cuba. In the Entomology. case of common bean, the use of varieties possessing high levels of resistance to BGYMV remains the main Current Status of strategy for virus control. Whitefly/Begomovirus Problems References The whitefly situation in Cuba worsened following the invasion of the Blanco, N.; Bencomo, I. 1978. Afluencia de la mosca blanca (Bemisia tabaci) new B biotype of B. tabaci. However, vector del virus del mosaico dorado the relatively low use of insecticides en plantaciones de frijol. Ciencias and the production of biological control de la Agricultura 2:39-45. agents in Cuba should maintain the whitefly problem at a manageable level. Brown, J. K. 1994. Current status of The Cuban provinces most affected by Bemisia tabaci as a plant pest and B. tabaci and the viruses it transmits virus vector in agro-ecosystems worldwide. FAO Plant Protect. Bull. have been Pinar del Río and Havana in 42:3-32. the western end of the island and Holguín and Las Tunas in the eastern Cancino, M.; Abouzid, A. M.; Morales, F. half of the country. The whitefly J.; Purcifull, D. E.; Polston J. E.; problem also has been severe in certain Hiebert, E. 1995. Generation and years in the provinces of Santiago and characterization of three Guantánamo in the eastern end of monoclonal antibodies useful in detecting and distinguishing Bean Cuba. Other provinces affected have golden mosaic virus isolates. been Matanzas, Ciego de Avila, Phytopathology 85:484-490.

235 Whiteflies and Whitefly-borne Viruses in the Tropics

Murguido, C.; Vázquez, L.; Gómez, O.; Ramos, P. L.; Guerra, O.; Peral, P.; Mateo, A. 1997. Informe sobre la Oramas, P.; Guevara, R. G.; Rivera- problemática mosca blanca- Bustamante, R. 1997. Taino tomato geminivirus. Instituto de mottle virus, a new bipartite Investigaciones de Sanidad Vegetal geminivirus from Cuba. Plant Dis. (INISAV), Boletín Técnico no. 5, 81:1095. La Habana, CU. 13 p. Rojas, M. R. 1992. Detection and Polston, J. E.; Anderson, P. K. 1997. The characterization of whitefly- emergence of whitefly-transmitted transmitted geminiviruses by the geminiviruses in tomato in the use of polymerase chain reaction. western hemisphere. Plant Dis. 81: University of Wisconsin, Madison, 1358-1369. USA.

236 Dominican Republic

CHAPTER 3.11 Dominican Republic

Francisco Morales*, Emigdio Gómez**, Augusto Villar** and Julio Nin***

Introduction hurricanes are a major threat every year (West and Augelli, 1977). Figure 1 shows the main agricultural regions affected by Geographical context whitefly-transmitted begomoviruses. The Dominican Republic comprises the eastern two-thirds of the island of Hispaniola. The country is generally The emergence of Bemisia tabaci mountainous, with the most prominent as a pest and virus vector range being the Central Highlands (elevation up to 3175 m), the highest Bemisia tabaci (Gennadius) was first point in the West Indies. The Constanza observed attacking common bean in Valley is a highly developed agricultural 1975. However, the presence of Bean area located in the Central Highlands golden yellow mosaic virus (BGYMV) in and is currently planted to high-value the Dominican Republic had been crops such as garlic (Allium sativum L.), noticed as early as the late 1960s onion (Allium cepa L.) and other (Schieber, 1970). By 1988, other crops horticultural crops. The Cibao Valley, in such as tomato (Lycopersicon the north-west, is one of the most fertile esculentum Mill.), melon (Cucumis melo agricultural areas in the country, where L.), eggplant (Solanum melongena L.), rice (Oryza sativa L.), maize (Zea mays cucumber (Cucumis sativus L. var. L.), common bean (Phaseolus vulgaris sativus) and watermelon (Citrullus L.), tobacco (Nicotiana tabacum L.) and lanatus [Thunb.] Matsum. & Nakai) were coffee (Coffea arabica L.) are produced. under attack from B. tabaci. In 1991, The western part of the country is dry biotype B of B. tabaci made its but rivers provide irrigation. San Juan appearance in the main horticultural de la Maguana is the main bean area regions of the Dominican Republic. At and the Azua Valley is the main the same time, the first begomovirus horticultural area in south-western from the Old World, Tomato yellow leaf Dominican Republic. The annual mean curl virus (TYLCV) was irresponsibly temperature is 25 °C and precipitation introduced into the Dominican Republic is 1346 mm. Tropical storms and and the Americas. Following the introduction of a new and more * Centro Internacional de Agricultura Tropical aggressive biotype of the whitefly vector (CIAT), Cali, Colombia. and an exotic begomovirus, crops such ** Programa Nacional de Manejo Integrado de as tomato and melon practically Plagas (MIP), Secretaría del Estado de Agricultura (SEA), Dominican Republic. disappeared, and the tomato processing *** Programa de Mejoramiento, SEA, Dominican plants had to close down in certain Republic. areas.

237 Whiteflies and Whitefly-borne Viruses in the Tropics

Monte Puerto Plata Cristi Espaillat Valverde María Trinidad Dajabon Salcedo Santiago Sánchez Rodríguez Santiago Duarte Samana

La Vega Sánchez Ramírez San Juan Monseñor Noube Monte Plata Hato El Seibo Mayor La Altagracia Peravia Distrito Baoruco Azua San Pedro de La Nacional Romana San Macoris Elías Cristóbal Pina

Barahona La Romana Dominican Republic

Figure 1. The main agricultural regions affected by whitefly-transmitted begomoviruses, Dominican Republic.

Advances in Biological presence of a begomovirus infecting Research tomatoes in the Dominican Republic was first noticed in 1992 and The first begomovirus of economic subsequently identified as TYLCV from importance to be characterized at the Israel (Polston et al., 1994; Polston and molecular level in the Dominican Anderson, 1997). The introduction of Republic was BGYMV. This virus was an exotic begomovirus from the Old isolated in the early 1990s by the World created a major pandemic senior author and later shown to be throughout the Caribbean at a cost to molecularly and serologically related to tomato growers and tomato the Guatemalan isolate of BGYMV processing plants worth millions of (Faria et al., 1994; Cancino et al., US dollars in yield/export losses 1995). The BGYMV isolate from the (Dupuy, 1998). Dominican Republic (San Juan de la Maguana) proved to be similar to At the beginning of this project, in BGYMV isolates from the Caribbean April 1997, common bean samples (Puerto Rica) and Central America were collected in San Juan de la (Guatemala). The BGYMV isolate from Maguana and tomato samples in Azua, the Dominican Republic was shown to in south-western Dominican Republic. be a different species when compared The common bean samples reacted with Bean golden mosaic virus (BGMV) with the broad-spectrum monoclonal from Brazil. BGYMV affects over antibody 3F7 as well as with the 20,000 ha of common bean in the monoclonal antibody prepared to the Dominican Republic each year, Guatemalan isolate of BGYMV. particularly in the south-western valley However, the tomato samples did not of San Juan de la Maguana. The react against the broad-spectrum

238 Dominican Republic monoclonal antibody (MAB) in these The common bean samples from tests. These tomato samples were then the valley of San Juan de la Maguana assayed by polymerase chain reaction behaved as expected for the BGYMV (PCR) using two sets of primers isolates previously detected in this (PTYIRV21/PTYIRC287 and PTYIRV21/ locality. The presence of BGYMV in the RTYC2C1814) to TYLCV provided by Constanza Valley is interesting, Dr. M. Nakhla, Plant Pathology because it is located at altitudes Department of the University of between 1000 and 1100 m, above the Wisconsin, Madison, WI. These primers altitudinal range (0-950 m) at which amplified two fragments of Mr 287 and B. tabaci is usually a problem. 1793 pb, respectively, demonstrating However, the incidence of the virus was the presence of TYLCV in the tomato moderate in the Constanza Valley. plants sampled in Azua.

In February 1998, six more tomato Socio-economic Analysis samples from Azua and one from the northern locality of Santiago were The agriculture of the Dominican assayed with similar results (Table 1), Republic was described in the 1970s as demonstrating the endemic nature of a mixture of traditional and modern TYLCV in the Dominican Republic. The cropping systems. Traditional three tomato samples (9-11) from the agriculture was equivalent to Centro de Investigaciones Aplicadas a “subsistence” agriculture and included Zonas Áridas (CIAZA), Azua, crops such as common bean, cassava corresponded to TYLCV-tolerant tomato (Manihot esculenta Crantz), rice, varieties (Gem-Pear, Gem-Star and sweetpotato (Ipomoea batatas [L.] UC82), which produce acceptably in Lam.), pigeon pea (Cajanus cajan [L.] this locality. Millsp.) and taro (Colocasia esculenta

Table 1. Analyses of selected plant samples from the Dominican Republic.

Sample Plant Localitya MAB-BSb MAB-GAc PTY1d TYLCVe

1 Common bean San Juan + + - nt 2 Common bean San Juan - - - - 3 Common bean Constanza + + - nt 4 Common bean Constanza + + - nt 5 Garlic Constanza nt nt + nt 6 Tomato Azua - nt nt + 7 Tomato Azua - nt nt + 8 Tomato Azua - nt nt + 9 Tomato CIAZA-Azua - nt nt + 10 Tomato CIAZA-Azua - nt nt + 11 Tomato CIAZA-Azua - nt nt + 12 Tomato Santiago - nt nt + 13 Tomato Santiago - nt nt + a. CIAZA, Centro de Investigaciones Aplicadas a Zonas Áridas. b. MAB-BS, a broad spectrum monoclonal antibody used to detect bi-partite begomoviruses. c. MAB-GA, a monoclonal antibody used to detect the original Middle American isolates of Bean golden yellow mosaic virus-Guatemala; and nt, no tests for these samples. d. PTYI, monoclonal antibodies to detect potyviruses; and nt, no tests for these samples. e. TYLCV, polymerase chain reaction detection of Tomato yellow leaf curl virus; and nt, no test for these samples.

239 Whiteflies and Whitefly-borne Viruses in the Tropics

[L.] Schott). Modern agriculture was the yield losses in the production of equivalent to the “plantation” industrial tomato for the two main agriculture going back to colonial production areas, Azua and Cibao, times, represented mostly by the between 1988 and 1995 (Alvarez and extensive cultivation of sugarcane Abud-Antún, 1995). This crisis was (Saccharum officinarum L.). In 1970, managed through the adoption of legal sugarcane was the main agro-industry measures that included a 3-month of the country, occupying over crop-free period to break the cycle of 150,000 ha. the whitefly vector. Additionally, virus- resistant tomato cultivars were In 1966, a Land Reform introduced. Programme began to be implemented in the Dominican Republic. Large-scale Table 2. Yield losses (%) caused by whitefly/ farming in the form of co-operatives begomovirus damage to industrial began to take place with the support of tomatoes in the main tomato the government, the financial producing regions of the Dominican Republic. contribution of international agencies —for example the Agency for Cropping Region season International Development (AID)—and Azua Cibao the technical support of third 1988-89 25 5 countries. The Sisal Project, for 1989-90 45 10 instance, was created at that time with 1990-91 40 15 the help of the Government of Israel. 1991-92 30 15 The objective of the project was to 1992-93 80 80 produce tomato under irrigation in the 1993-94 95 50 valley of Azua. The irrigation system 1994-95 20 15 was initiated in 1970 with the construction of the Sabana Yegua Dam to irrigate some 18,000 ha. Thus, the Following the implementation of Dominican Republic was one of the legal measures against B. tabaci, the first countries in Latin America to look melon industry made a comeback. The into the possibility of producing non- case of common bean was not traditional, high-value crops for export promising, because this crop was being such as tomato, chilli (Capsicum displaced from the traditional bean- annuum L. var. annuum), melon, growing areas such as the valley of San watermelon and eggplant. In 1970, Juan de la Maguana by other crops 1600 ha of tomato were planted in the without whitefly problems and in Dominican Republic. In the late 1990s, higher demand, including cassava and the tomato industry covered 8000 ha, sweetpotato. In 1984, there were employed over 6500 small-scale 63,000 ha of red-seeded bean in the farmers (farms averaging less than country and in 1994 the area planted 2.5 ha) and created jobs for to common bean is only 36,000 ha, 90,000 people per crop cycle (Dupuy, almost half of the area planted 1998). 15 years ago. As a result, the country has had to import common bean, The arrival of TYLCV from Israel usually grain types of little demand in and biotype B of B. tabaci brought the the country, considering that red- thriving tomato industry to a halt, seeded bean is the main food staple in causing economic losses in excess of the diet of the Dominicans (Fundación 30 million US dollars. Table 2 shows de Desarrollo Agropecuario, 1995).

240 Dominican Republic

Strengthened Research utilization of monoclonal and Capacity polyclonal antibodies and financed by the project to attend the International Considering that the two main Workshop on Bemisia and horticultural regions affected by Geminiviruses, held in San Juan, whiteflies and whitefly-transmitted Puerto Rico, June 1998. Here they geminiviruses in the Dominican presented the novel approach adopted Republic are Azua and Cibao, two in the Dominican Republic to control Agricultural Engineers were contacted whitefly/begomovirus damage by to collaborate in this project. They were regulatory measures designed to break Augusto Villar of the Secretaría del the continuous populational cycle of Estado de Agricultura (SEA)-CIAZA, B. tabaci in tropical environments. Azua, and Emigdio Gómez of SEA- Programa Nacional de Manejo The project also supported the Integrado de Plagas (MIP), N-N.W. activities of several researchers on Cibao. These national programme horticultural crops in the Constanza scientists were brought to the Centro Valley, in the central region of the Internacional de Agricultura Tropical country, in collaboration with the (CIAT) for training in the identification horticultural research network, La Red of B. tabaci biotypes and diagnosis of Colaborativa de Investigación y viruses affecting crops of socio- Desarrollo de Hortalizas para América economic importance in their Central, Panamá y República respective agricultural regions. Table 3 Dominicana (REDCAHOR). Two shows the results obtained by these conferences were delivered on the two researchers at CIAT with the Tropical Whitefly Integrated Pest whitefly samples they collected in the Management (TWF-IPM) Project and on Dominican Republic. the breeding and selection of horticultural crops for their resistance The two researchers were trained to whitefly-transmitted begomoviruses. also in the identification of The CIAT Bean Project is collaborating geminiviruses and other plant viruses with the Dominican Republic to breed using the enzyme-linked Dominican bean cultivars, mainly the immunosorbent assay (ELISA) Pompadour types, for resistance to technique in conjunction with the BGYMV.

Table 3. Results of the analyses of Bemisia spp. biotypes found in the Dominican Republic on different economically important crops.

Locality Crop No. samples Whitefly

Azua-CIAZAa Eggplant 3 Bemisia tuberculata Azua-Estebanía Tomato 9 Bemisia tabaci-B Barahona Eggplant 3 Bemisia tuberculata Azua-Arroyo Salado Eggplant 3 Bemisia tuberculata San José de Ocoa Tomato 6 Bemisia tabaci-B Santiago Okra 3 Bemisia tabaci-B Santiago Cucumber 3 Bemisia tabaci-B Santiago Eggplant 3 Bemisia tabaci-B Montecristi Melon 6 Bemisia tabaci-B Santiago Tomato 3 Bemisia tabaci-B Santiago Okra 3 Bemisia tabaci-B a. CIAZA, Centro de Investigaciones Aplicadas a Zonas Áridas.

241 Whiteflies and Whitefly-borne Viruses in the Tropics

Current Situation of Dupuy, J. 1998. Dealing with whiteflies Whitefly/Geminivirus and geminiviruses in the Dominican Republic. Special Report. Barceló Problems Industries, Santo Domingo, DO. 12 p. The Dominican Republic constitutes a unique example of a country that Faria, J. C.; Gilbertson, R. L.; Hanson, S. implemented legal measures to minimize F.; Morales, F. J.; Ahlquist, P.; yield losses due to the activity of B. Loniello, A. O.; Maxwell, D. P. 1994. tabaci both as a pest and virus vector. Bean golden mosaic geminivirus type II isolates from the Dominican Nevertheless, pesticide use remains Republic and Guatemala: high in most agricultural regions of the Nucleotide sequences, infectious country, and the existing cropping pseudorecombinants, and systems favour the reproduction of phylogenetic relationships. B. tabaci. Currently, the situation in the Phytopathology 84:321-329. lowlands and highlands remains under partial control, and some crops that Fundación de Desarrollo Agropecuario. 1995. Cultivo de habichuela. had been taken out of production such Boletín Técnico no. 2. Santo as melon are being planted again. Domingo, DO. 38 p. Another crop and good host to the whitefly B. tabaci, tobacco, also has Navas-Castillo, J.; Sánchez-Campos, S.; been increasing its area in the country. Díaz, J. A. 1999. Tomato yellow leaf A recent viral epidemic in tobacco in the curl virus-Is causes a novel disease Cibao region was diagnosed at CIAT as of common bean and severe epidemics in tomato in Spain. Plant being caused by Tobacco mosaic virus. Dis. 83:29-32.

In the case of tomato, the Polston, J. E.; Bois, D.; Serra, C. A.; cultivation of varieties resistant to Concepción, S. 1994. First report of TYLCV should contribute to the a tomato yellow leaf curl-like recovery of the production of industrial geminivirus in the western tomato plantings. In the case of hemisphere. Plant Dis. 78:831. common bean, however, we have Polston, J. E.; Anderson, P. K. 1997. The noticed a marked reduction in the area emergence of whitefly-transmitted planted, because of the susceptibility of geminiviruses in tomato in the most local varieties to BGYMV. western hemisphere. Plant Dis. Moreover, TYLCV has been reported to 81:1358-1369. attack common bean in other parts of the world (Navas-Castillo et al., 1999). Schieber, E. 1970. Enfermedades del frijol (Phaseolus vulgaris L.) en la República Dominicana. Turrialba 20:20-23. References West, R. C.; Augelli, J. P. 1977. Middle Alvarez, P. A.; Abud-Antún, A. 1995. America: Its lands and peoples. Reporte de la República Dominicana. Second edition. Prentice-Hall, NJ, CEIBA 36:39-47. USA. 494 p.

Cancino, M.; Abouzid, A. M.; Morales, F. J.; Purcifull, D. E.; Poston, J. E.; Hiebert, E. 1995. Generation and characterization of three monoclonal antibodies useful in detecting and distinguishing Bean golden mosaic virus isolates. Phytopathology 85:484-490.

242 Reproductive Crop Hosts of Bemisia tabaci

CHAPTER 3.12 Reproductive Crop Hosts of Bemisia tabaci (Gennadius) in Latin America and the Caribbean

Pamela K. Anderson*, Avas Hamon**, Pilar Hernández*** and Jon Martinψ

Introduction Crop Hosts of Bemisia tabaci The sweetpotato whitefly Bemisia tabaci (Gennadius) is a vector of at least The most reliable sources of host 30 begomoviruses (Geminiviridae: plant records for B. tabaci are Begomovirus) in Latin America and the published taxonomic reports and Caribbean (Morales and Anderson, unpublished records from museum 2001). As is often the case for insect collections. The first author has vector-host relations, many of the reviewed the white and grey literature crops that begomoviruses affect are not as well as the arthropod collections reproductive hosts for B. tabaci. A first from the United States National critical step in vector entomology, then, Museum (USNM), the Florida State is to answer the question. “Where is Collection of Arthropods (FSCA), the B. tabaci reproducing?” British Museum of Natural History (BMNH), and the Escuela Agrícola Crop biomass can provide excellent Panamericana (EAP) in Honduras. The breeding and feeding sources and thus EAP (1992) references are give rise to large populations of unpublished records from Latin B. tabaci (Byrne et al., 1991). We argue America, collected and identified by here that the first step to R. Caballero as the basis for understanding regional population development of several taxonomic dynamics as the basis for area-wide keys (Caballero, 1992). Table 1 also management is to elucidate which records B. tabaci host records crops B. tabaci utilizes as reproductive generated by the national program hosts and define their relative (national agricultural research importance. systems [NARS]) teams as part of the extensive diagnostic surveys for Phase 1 of the Tropical Whitefly Integrated Pest Management Project * Former coordinator for Phase 1 of the (TWF-IPM). P. Hernández identified Tropical Whitefly IPM Project; currently these specimens and A. Hamon Research Director at the Centro Internacional de la Papa (CIP), Lima, Peru. verified them. Verified specimens are ** Florida State Collection of Arthropods deposited in the insect museum at (FSCA), Gainesville, Florida, USA. the Centro Internacional de *** Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. Agricultura Tropical (CIAT) in Cali, ψ British Museum of Natural History, London, Colombia, and the FSCA in UK. Gainesville, Florida, USA.

243 Whiteflies and Whitefly-borne Viruses in the Tropics

Based upon the reliable published B. tabaci is utilizing at least taxonomic reports, unpublished 50 cultivated species as reproductive museum records and the verified hosts in Latin America and the specimen collected during Phase 1 of Caribbean (Table 1). the TWF-IPM, we conclude that

Table 1. Known crop hosts of Bemisia tabaci in Latin America and the Caribbean.

Crop Country Referencea Latin name Common name

Abelmoschus esculentus (L.) Moench Okra Cuba Vázquez et al. (1996) [= Hibiscus esculentus L.] Dom. Rep. TWF-IPM Project El Salvador USNM (1967), Lipes (1968), WF-IPM Project Guatemala TWF-IPM Project Trinidad BMNH (1989) Venezuela Arnal et al. (1993b) Annona muricata L. Soursop Cuba Vázquez et al. (1996) Arachis hypogaea L. Peanut Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Beta vulgaris L. Beet Cuba Vázquez et al. (1996) Beta vulgaris subsp cicla (L.) W. Koch Swiss chard Cuba Vázquez et al. (1996) Brasica rapas L. subsp. rapa Turnip Cuba Vázquez et al. (1996) Brassica oleracea L. var. capitata L. Cabbage Cuba Vázquez et al. (1995) Brassica oleracea L. var. gemminifera DC Brussel sprouts Venezuela Arnal et al. (1993b) Brassica oleracea L. var. gongylodes L. Kohl-rabi Cuba Vázquez et al. (1995) Cajanas cajan (L.) Millsp. Pigeon pea Cuba Vázquez et al. (1996) Canavalia ensiformis (L.) DC. Jack bean, Cuba Vázquez et al. (1996) sword bean Capsicum annuum L. var. annuum Bell pepper Belize EAP (1992) Costa Rica TWF-IPM Project Cuba Vázquez et al. (1996) Honduras EAP (1992) Panama TWF-IPM Project Venezuela Arnal et al. (1993b) Capsicum spp. Chilli peppers Costa Rica TWF-IPM Project Cuba Vázquez et al. (1996) El Salvador TWF-IPM Project Guatemala TWF-IPM Project Honduras TWF-IPM Project Nicaragua EAP (1992) Venezuela Arnal et al. (1993b) Carica papaya L. Papaya Cuba Vázquez et al. (1996) Guatemala TWF-IPM Project Cicer arietinum L. Chick-pea Cuba Vázquez et al. (1996) Citrullus lanatus (Thunb.) Matsum. & Nakai Watermelon Belize EAP (1992) [= Citrullus vulgaris Schrad.] Colombia TWF-IPM Project Costa Rica TWF-IPM Project Cuba Vázquez et al. (1995) Ecuador TWF-IPM Project El Salvador TWF-IPM Project Guatemala TWF-IPM Project Honduras EAP (1992) Panama TWF-IPM Project Venezuela Arnal et al. (1993b) Cucumis melo L. Melon Belize EAP (1992) Colombia TWF-IPM Project Costa Rica TWF-IPM Project Dom. Rep. TWF-IPM Project Ecuador TWF-IPM Project El Salvador TWF-IPM Project (Continued)

244 Reproductive Crop Hosts of Bemisia tabaci

Table 1. (Continued.)

Crop Country Referencea Latin name Common name

Guatemala EAP (1992), TWF-IPM Project Honduras EAP (1992) Venezuela Arnal et al. (1993a; 1993b), TWF-IPM Project Cucumis sativus L. Cucumber Colombia TWF-IPM Project Costa Rica TWF-IPM Project Cuba Vázquez et al. (1996) Dom. Rep. TWF-IPM Project El Salvador TWF-IPM Project Guatemala TWF-IPM Project Honduras EAP (1992), TWF-IPM Project Venezuela Arnal et al. (1993b), TWF-IPM Project Cucurbita argyrosperma C. Huber Pipián El Salvador TWF-IPM Project subsp. argyrosperma Cucurbita maxima Duch. Ex Lam. Squash Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b), TWF-IPM Project Cucurbita moschata (Duch. ex Lam.) Ayote Colombia TWF-IPM Project Duch. ex Poir. El Salvador TWF-IPM Project Honduras EAP (1992) Panama TWF-IPM Project Cucurbita pepo L. Squash Ecuador TWF-IPM Project Nicaragua USNM (1958) Venezuela Arnal et al. (1993b) Euphorbia pulcherrima Willd. Ex Klotzch Poinsettia El Salvador USNM (1967) Fernaldia pandurata (A. DC.) Woodson Loroco El Salvador TWF-IPM Project Glycine max (L.) Merr. Soybean Argentina TWF-IPM Project Brazil BMNH (1974) Colombia EAP (1992), TWF-IPM Project Cuba Vázquez et al. (1996) Ecuador TWF-IPM Project Honduras EAP (1992) Venezuela Arnal et al. (1993a; 1993b) Gossypium hirsutum L. Cotton Argentina Viscarrret and Botto (1996) Barbados BMNH (1982) Colombia EAP (1992), TWF-IPM Project El Salvador USNM (1964; 1965; 1967; 1969) Guatemala EAP (1992) Honduras USNM (1966) Nicaragua USNM (1951; 1965; 1978) Venezuela Arnal et al. (1993b) Helianthus annuus L. Sunflower Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Ipomoea batatas (L.) Lam. Sweetpotato Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Lactuca sativa L. Lettuce Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Lycopersicon esculentum Mill. Tomato Belize EAP (1992) Colombia EAP (1992), TWF-IPM Project (Continued)

245 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 1. (Continued.)

Crop Country Referencea Latin name Common name Costa Rica TWF-IPM Project Cuba Vázquez et al. (1996) Dom. Rep. EAP (1992), TWF-IPM Project El Salvador USNM (1967), TWF-IPM Project Guatemala EAP (1992), TWF-IPM Project Honduras EAP (1992), TWF-IPM Project Mexico BMNH (1992) Nicaragua USNM (1952), TWF-IPM Project Panama TWF-IPM Project Puerto Rico BMNH (1988) Trinidad BMNH (1990) Venezuela Arnal et al. (1993a; 1993b), TWF-IPM Project Mangifera indica L. Mango Cuba Vázquez et al. (1996) Manihot esculenta Crantz Cassava Colombia EAP (1992), TWF-IPM Project Cuba Vázquez et al. (1996) Dom. Rep. TWF-IPM Project Ecuador TWF-IPM Project Venezuela Arnal et al. (1993b) Musa × paradisiacal L. Banana/plantain Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Nicotiana tabacum L. Tobacco Cuba Vázquez et al. (1996) Guatemala TWF-IPM Project Nicaragua USNM (1951) Venezuela Arnal et al. (1993b) Ocimum basilicum L. Basil Cuba Vázquez et al. (1996) Passiflora edulis Sims Passionfruit Venezuela Arnal et al. (1993b) Persea americana Mill. Avocado Cuba Vázquez et al. (1996) Guatemala USNM (1954) Venezuela Arnal et al. (1993b) Phaseolus acutifolius A. Gray Tepary bean El Salvador TWF-IPM Project var. acutifolius Phaseolus lunatus L. Lima bean Brazil BMNH (1975) Venezuela Arnal et al. (1993b) Phaseolus vulgaris L. Common bean Argentina USNM (1982), TWF-IPM Project Colombia USNM (1974), TWF-IPM Project Costa Rica TWF-IPM Project Cuba Vázquez et al. (1996) Dom. Rep. BMNH (1979); TWF-IPM Project Ecuador TWF-IPM Project El Salvador USNM (1964); TWF-IPM Project Phaseolus vulgaris L. Common bean Guatemala USNM (1973); TWF-IPM Project Honduras EAP (1992); TWF-IPM Project Nicaragua USNM (1974); TWF-IPM Project Panama TWF-IPM Project Venezuela Arnal et al. (1993a; 1993b)

(Continued)

246 Reproductive Crop Hosts of Bemisia tabaci

Table 1. (Continued.)

Crop Country Referencea Latin name Common name Psidium guajava L. Guava Cuba Vázquez et al. (1996) Nicaragua USNM (1978) Raphanus sativus L. Radish Cuba Vázquez et al. (1995) Sesamum indicum L. Sesame Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Solanum melongena L. Eggplant Colombia TWF-IPM Project Cuba Vázquez et al. (1996) Dom. Rep. EAP (1992), TWF-IPM Project Ecuador TWF-IPM Project Guatemala TWF-IPM Project Mexico TWF-IPM Project Puerto Rico BMNH (1987) Trinidad BMNH (1988) Venezuela Arnal et al. (1993b) Solanum tuberosum L. Potato Argentina USNM (1982) Cuba Vázquez et al. (1996) Honduras EAP (1992) Nicaragua EAP (1992) Venezuela Arnal et al. (1993a; 1993b) Spondias purpurea L. Spanish plum Cuba Vázquez et al. (1996) Vicia faba L. Broad bean Cuba Vázquez et al. (1996) Vigna unguiculata (L.) Walp. Cowpea Cuba Vázquez et al. (1996) Venezuela Arnal et al. (1993b) Vigna unguiculata subsp. Yard-long bean Cuba Vázquez et al. (1996) sesquipedalis (L.) Verdc. Xanthosoma sagittifolium (L.) Schott Malanga Cuba Vázquez et al. (1996) a. For references, see list at end of chapter. BMNH, British Museum of Natural History, data from various dates; EAP, Escuela Agrícola Panamericana, Honduras, unpublished data; USNM, United States National Museum, data from various dates; TWF-IPM Project, Tropical Whitefly-Integrated Pest Management Project.

Relative Importance of host race is a population of a species Bemisia tabaci on that is partially reproductively Different Crop Hosts isolated from other non-specific, sympatric sister populations as a direct consequence of adaptation to a B. tabaci is cosmopolitan in specific host or habitat (Diehl and distribution, highly polyphagous and Bush, 1984; Bush, 1994). Host races characterized by intercrop movement are one category of biotypes, that is, (Butler et al., 1986). Based on two or more morphologically similar identified museum specimens, Mound or indistinguishable taxa that differ and Halsey (1978) listed 317 plant from one another in a biologically species as host plants for B. tabaci, significant way such as host worldwide. However, sufficient preference, emergence time or some empirical and experimental evidence other ecological or behavioral trait exists to indicate B. tabaci populations (Bush, 1994). do not reproduce equally among all potential crop hosts in any given In any given geographical area, geographical region. Rather, B. tabaci beyond more precisely defining the appears to develop strong host range of reproductive crop hosts, it is associations, resulting in host races. A necessary to define the relative

247 Whiteflies and Whitefly-borne Viruses in the Tropics importance, or reproductive potential, measures on a per ground unit basis of the host plants, that is, to quantify are necessary for meaningful the average number of B. tabaci comparisons. Recently S. Naranjo and produced per unit area. For example, L. Cañas (United States Department of intensive surveys carried out in Agriculture [USDA]- Agricultural Venezuela (Arnal et al., 1993a; 1993b) Research Service [ARS]) have and Cuba (Vázquez et al., 1995; 1996) developed a protocol to quantify each identified 27 and 39reproductive B. tabaci abundance for a range of crop hosts for B. tabaci respectively crop plants within the same area on a (Table 1). The next step is to identify per ground unit basis. This protocol is the relative importance of each crop currently being tested in Arizona as a reproductive host. (Cañas, Naranjo and Ellis, in preparation) and in El Salvador Caballero and Nolasco (1995) (Serrano and Anderson, in made 289 collections of B. tabaci from preparation). Such quantification will economically important crop hosts in allow us to compare the relative Belize, Costa Rica, El Salvador, importance of B. tabaci reproductive Guatemala, Honduras, Nicaragua and hosts within and among zones in a Panama. They quantified B. tabaci country, and among different abundance for each crop utilizing countries; to monitor fluctuations in B. tabaci density/leaf as an indicator the B. tabaci populations on the same of abundance (Table 2). crop in the same zone from year to year; and to identify newly emerging However, comparisons of whitefly reproductive hosts for B. tabaci. abundance on different crops is problematic because of variation in We cannot effectively implement architecture, growth habits, foliage crop protection tactics aimed at density and distribution of B. tabaci source reduction within an area-wide within the plant, among different management program if we cannot crops and among different varieties prioritize and target the principal within the same crop. Density reproductive sources for B. tabaci.

Table 2. Relative importance of Bemisia tabaci crop hosts in Central America, based on leaf samples.

Rankinga Crop Common name

5 Gossypium hirsutum L. Cotton Citrillus lanatus (Thumb.) Matsum. & Nakai Watermelon Cucurbita pepo L. Squash 4 Lycopersicon esculentum Mill. Tomato Cucmus melo L. Melon Capsicum annuum L. Sweet pepper 3 Nicotiana tabacum L. Tobacco Solanum melongena L. Eggplant Phaseolus vulgaris L. Common bean 2 Glycine max L. (Merr.) Soybean a. Rank 5, up to 100 nymphs per leaf, in all countries surveyed; rank 4, up to 100 nymphs per leaf, in certain countries; rank 3, up to 50 nymphs per leaf, in all countries; rank 2, up to 20 nymphs per leaf, in all countries; and rank 1, less than 10 nymphs per leaf, in all countries.

248 Reproductive Crop Hosts of Bemisia tabaci

Conclusions and Byrne, D. N.; Bellows, T. S.; Parrella, M. Recommendations P. 1991. Whiteflies in agricultural systems. In: Gerling, D. (ed.). Whiteflies: Their bionomics, pest After more than a decade of epidemic status and management. Intercept, events in Mexico, Central America and Andover, Hants, GB. p. 227-261. the Caribbean, we still do not have a clear enough understanding of the Caballero, R. 1992. Whiteflies range of reproductive crop hosts for (Homoptera: Aleyrodidae) from B. tabaci. Nor do we understand the Central America and Colombia relative importance of the B. tabaci including slide-mounted pupal and field keys for identification, field reproductive hosts within a zone, for characteristics, hosts, disbribution, purposes of area-wide management. natural enemies, and economic The question, “where is B. tabaci importance. M.Sc. thesis, Kansas reproducing?”, continues to be a basic State University, USA. 201 p. research priority for vector entomology in the region. Caballero, R.; Nolasco, C. 1995. Determinación de especies de moscas blancas, biotipos de Bemisia tabaci (Genn.), malezas References hospederas y geminivirus. Programa regional de reforzamiento Arnal, E.; Russell, L. M.; Debrot, E.; de la investigación agronómica Ramos, F.; Cermeli, M.; Marcano, sobre los granos básicos en Centro R.; Montagne, A. 1993a. Lista de América (ITR 9305-0793). 11 p. moscas blancas (Homoptera: Aleyrodidae) y sus plantas Diehl, S. R.; Bush, G. 1984. An hospederas en Venezuela. Fla. evolutionary and applied Entomol. 76:365-381. perspective of insect biotypes. Ann. Rev. Entomol. 29:471-504. Arnal, E.; Ramos, F.; Debrot, E. 1993b. Plantas hospederas de la mosca Lipes, J. E. 1968. El Salvador: blanca Bemisia tabaci (Gennadius) Occurrence of the banded-wing en Venezuela. Agron. Trop. whitefly, Trialeurodes abutilonea. (Venezuela) 43:267-286. FAO Plant Prot. Bull. 16:32.

Bush, G. L. 1994. Bemisia tabaci: Biotype Morales, F. J.; Anderson, P. K. 2001. The or species complex? In: Rojas, A.; emergence and dissemination of Varela, G.; Valle, N.; Chavarria, J. whitefly-transmitted geminiviruses A. (eds.). Memoria, II Taller in Latin America. Arch. Virol. Latinoamericano y del Caribe sobre 146 (3):415-441. moscas blancas y geminivirus, 20-22 octubre 1993, Organismo Mound, L. A.; Halsey, S. H. 1978. Internacional Regional de Sanidad Whitefly of the world: A systematic Agropecuaria (OIRSA), Managua, catalogue of the Aleyrodidae NI. p. 17-24. (Homoptera) with host plant and natural enemy data. John Wiley, Butler, G. D. Jr.; Henneberry, T. J.; Chichester, GB. 340 p. Hutchinson, W. D. 1986. Biology, sampling and population dynamics Vázquez, L. L.; de la Iglesia, M.; López, D.; of Bemisia tabaci. Agric. Zool. Rev. Jiménez, R.; Mateo, A.; Vera, E. R. 1:167-195. 1995. Moscas blancas (Homoptera: Aleyrodidae) detectadas en los principales cultivos agrícolas de Cuba. MIP (Costa Rica) 36:18-21.

249 Whiteflies and Whitefly-borne Viruses in the Tropics

Vázquez, L. L.; Jiménez, R.; de la Iglesia, Viscarret, M. M.; Botto, E. N. 1996. M.; Mateo, A.; Borges, M. 1996. Descripción e identificación de Plantas hospederas de Bemisia Trialeurodes vaporariorum tabaci (Homoptera: Aleyrodidae) en (Westwood) y Bemisia tabaci Cuba. Rev. Biol. Trop. 44(3): (Gennadius) (Hemiptera, 143-148. Homoptera: Aleyrodidae). Rev. Chilena Ent. 23:51-58.

250 Biotypes of Bemisia tabaci

CHAPTER 3.13 Using Molecular Techniques to Analyse Whitefly Species and Biotypes in Latin America

Lee Calvert*, Natalia Villarreal* and Donald Frohlich**

Introduction starch—or polyacrylamide—gel electrophoresis have been in use for When identifying and describing insect several decades and are useful for taxa, morphology has been used processing large numbers of samples historically to separate species. Among relatively inexpensively compared to many groups of insects, however, DNA sequencing. However, because the morphological characters can vary with technique relies on detecting enzymes, respect to environmental factors within samples must be kept live or frozen in a single species, or be so convergent order to preserve activity. The and cryptic among closely related technique is also less sensitive than species as to be of limited usefulness. DNA approaches because the Under such conditions, studies of their underlying nucleic acid variability is biology and molecular profiles become usually masked. essential to defining species and characterizing populations. At a When, in the early 1980s, changes molecular level, protein and DNA in whitefly populations and associated polymorphisms can be combined with begomovirus infections were first studies of biological characteristics by noticed, protein polymorphisms were using one of four experimental or employed to investigate natural technological approaches: populations of B. tabaci. Differences in electrophoresis of allozymes, analysis esterase isozyme patterns were used to of randomly amplified polymorphic describe two biotypes in the Americas: DNAs (RAPDs) and nucleic acid a now-known-to-be native form, or sequence comparisons of nuclear or biotype A, and a second form, biotype mitochondrial DNA markers. Here, we B, which exhibited high population review the application of molecular density, wide host plant range, approaches to characterizing whitefly relatively high insecticide resistance Bemisia tabaci (Gennadius) populations and was capable of inducing “silverleaf” and biotypes in Latin America. symptoms on some plants (Brown et al., 1995). B. tabaci biotype A was Each method has its own predominant in most regions of the characteristic advantages and Americas but many of these disadvantages. Isozyme analyses using populations now have been displaced by B. tabaci biotype B. Other reports suggested that there might be * Centro Internacional de Agricultura Tropical additional biotypes in Latin America (CIAT), Cali, Colombia. (Wool et al., 1994). Eighteen other ** University of St. Thomas, Houston, TX, USA. biotypes from throughout the world

251 Whiteflies and Whitefly-borne Viruses in the Tropics have been described based on esterase that anneal randomly in the genome banding patterns alone (Brown et al., and produce characteristic banding 1995). Reliable morphological markers, patterns of collections of PCR products which do not vary with ecological when run on an agarose gel (RAPDs). parameters (Mound, 1963; David and Advantages here include the low cost Ananthakrishnan, 1976; Mohanty and and capability of processing large Basu, 1986) and which can distinguish numbers of samples without the between biotypes, are not known necessity of cloning or sequencing PCR (Rosell et al., 1997). products as well as no requirement for detailed sequence knowledge about the DNA sequence comparisons can be genome. Disadvantages include the fact made between individuals or that repeatability can be difficult, and populations using polymerase chain the polymorphisms may be hard to reaction (PCR) generated sequences distinguish from PCR artefacts. That is, from several markers in either the template preparation and amplification nuclear or mitochondrial genome. The conditions must be consistent and latter offers some advantages because tightly controlled. Interpretation of it is maternally inherited and non- negative data (absence of bands) also recombining and contains known and may have numerous explanations. predictable gene sequences for which general insect primers have been Perring et al. (1993) used RAPD designed. Some sequences vary enough patterns in combination with allozyme to be useful for population level frequency data and mating and analyses (e.g., the control region, parts behavioural studies to elevate the B of cytochrome oxidases I-III), while biotype to new species status, Bemisia others evolve slowly enough to be more argentifolii Bellows & Perring. In a appropriate to analyses above the detailed study of 20 RAPD primers, species level (e.g., conserved parts of however, Gawel and Bartlett (1993) cytochrome oxidase I [COI], 12S and concluded that the technique was not parts of the 16S r-DNA). However, as useful for clarifying taxon status in the with some nuclear markers, case of whiteflies. Campbell et al. comparisons between mitochondrial (1994) were the first to look at a sequences may reflect local gene nuclear DNA sequence, albeit highly evolution and not population or species conserved, and showed that only a evolution. Nuclear data offer a wide single unique nucleotide difference lies range of neutral markers but can be between B. tabaci biotypes A and B in particularly difficult to design PCR the 18S gene. More recently, Frohlich primers for specific targets in poorly et al. (1999) used variable portions of known or unknown genomes (e.g., the mitochondrial COI gene and introns). Thus, many studies have used 16S rDNA sequence to evaluate variable regions surrounded by B. tabaci populations from four relatively conserved sequences (e.g., continents, and concluded that biotype internal transcribed spacers or ITS of B is a recent introduction from the Old the rDNA genes) for population/species World to the Americas. De Barro et al. analyses, or more conserved sequences (2000), using ribosomal ITS 1, reached (e.g., 18S rDNA) for higher order the same conclusion and cautioned studies. that if the status of the B biotype were to remain species novum, a taxonomic An alternative approach to specific review of all of the biotypes of B. tabaci nuclear markers involves the use of would have to be made. short, random sequence PCR primers

252 Biotypes of Bemisia tabaci

For this study of the whiteflies different evolutionary assumptions, data affecting crops of economic importance, were evaluated by parsimony, neighbour a variable portion of the 16S joining and maximum-likelihood. All mitochondrial ribosomal DNA was analyses were performed with PAUP, chosen for evolutionary analysis. It is version 4.0b2, for Macintosh (Swofford, more variable than the 18S mt DNA but 1999). For parsimony, the branch-and- less so than the COI gene. This allows bound method was used (characters both a comparison of distinct genera as unordered, equal weight). Bootstrapping well as closely related biotypes. was performed with the branch-and- bound option for 2000 replicates (stepwise sequence addition, tree- Methods and Materials bisection-reconnection [TBR], MulTrees option). For neighbour joining, distances were calculated using the Kimura 2 PCR, cloning and sequence parameter model. Maximum-likelihood analysis of a region of the 16S trees were constructed with a mitochondrial DNA transition/transversion ratio of 2.0 by The primers 4119 heuristic search (100 replicates, random (5’ CGCCTGTTTAACAAAAACAT) and addition sequence, MulTrees, TBR) 4118 (5’ CCGGTCTGAACTCAG (Swofford, 1999). ATCACGT 3’) were used to amplify a region of the 16S mitochondrial DNA Whitefly survey (Xiong and Kocher, 1991). The PCR The whiteflies that were identified by reaction conditions were 30 cycles of RAPD PCR analysis were collected and 1 min at 95 °C, 50 s at 50 °C and processed.1 50 s at 72 °C, and in the last cycle the 72 °C reaction was extended to 10 min. RAPD PCR analysis The products were purified using the Total DNA was isolated from individual WizardTM PCR purification columns whiteflies using a method developed for (Promega, WI, USA) and were visualized plants (Gilbertson et al., 1991). The DNA by agarose gel electrophoresis. The PCR was amplified in a PCR. The Operon products were cloned into the plasmid primers F2 (5’GAGGATCCCT3’), PCR script amp SK(+)TM (Stratagene, F12 (5’ACGGTACCAG3’), H9 La Joya, CA, USA). Using the ABI dye (5’TGTAGCTGGG3’) and H16 terminator kit, the sequences were (5’TCTCAGCTGG3’) (De Barro and determined in an automated sequencer Driver, 1997) were tested for their using the dideoxynucleotide chain efficacy to distinguish whiteflies in Latin termination procedure (Sanger et al., America. The reaction conditions for the 1977). first cycle were 5 min at 94 °C, 2 min at Phylogenetic analyses 40 °C and 3 min at 72 °C. This was followed with 39 cycles of 1 min at Phylogenetic analyses were done with 94 °C, 1.5 min at 40 °C, and multiple individuals within populations. DNA sequences were aligned using the ClustalW algorithm (Thompson et al., 1994) by the ClustalW 1.7 program 1. For further information on specific (BCM Search Launcher at the Human methodologies used to produce the results Genome Centre, Baylor College of published in this book, please contact the information and communication assistant of Medicine, Houston, TX, USA). Because the Project (www.tropicalwhiteflyipmproject. different tree building algorithms make cgiar.org).

253 Whiteflies and Whitefly-borne Viruses in the Tropics

2 min at 72 °C. The PCR products were analysis of various B. tabaci, those of run in agarose gels stained with biotype A were grouped together and ethidium bromide and visualized using there was at least 97% identity UV light. with a maximum mean distance of 0.02 (Table 1) (Calvert et al., 2001). The B. tabaci biotype A is widespread Results throughout the region of the whitefly survey.

Analysis of selected whiteflies In the distance analysis, the Israel in Latin America B. tabaci isolate (Genbank accession: A region of the 16S mitochondrial DNA AF110717) was 98.8% identical with of B. tabaci biotypes A and biotype B, the B. tabaci biotype B isolates from B. tuberculata, Aleurotrachelus socialis Colombia (Table 1). Comparisons were (Bondar), Trialeurodes vaporariorum also made with individuals from (Westwood) and T. variabilis Colombia, Arizona, Israel and Yemen (Quaintance) were compared by and all were at least 98.2% identical parsimony and distance analysis using (data not shown). Given the rapid Phylip version 3.57 (Felsenstein, 1993). spread of the B biotype, the lack of At least two independent clones from diversity between populations in each whitefly population of the Arizona USA and Colombia was 3’ region of the mitochondrial 16S gene expected. This gives additional evidence were prepared and sequenced. Included that B. tabaci biotype B was recently in the comparison were sequence data introduced into the Americas from the of the mitochondrial 16S gene from region of the Middle East. B. tabaci biotype A of Arizona (Genbank accession: AF110722), Costa Rica A representative from each species (Genbank accession: AF110715) and or biotype was used in the analysis to Puerto Rico (Genbank accession: find the most parsimonious tree AF110719) (Frohlich et al., 1999). In the (Figure 1). B. tabaci biotypes A and

Table 1. Mean distances for a 3’ region of mitochondrial 16S ribosomal gene in fifteen individual whiteflies representing different species and populations.

Whiteflya 123456789101 112131415

1 B. tabaci B Sucre 0.00 0.01 0.01 0.01 0.10 0.10 0.09 0.10 0.10 0.22 0.35 0.35 0.37 0.37 0.36 2 B. tabaci B CT cass 0.00 0.00 0.00 0.08 0.09 0.08 0.09 0.09 0.21 0.34 0.34 0.36 0.36 0.35 3 B. tabaci B CT bn 0.00 0.00 0.09 0.10 0.09 0.09 0.09 0.21 0.34 0.35 0.37 0.36 0.35 4 B. tabaci Israel 0.00 0.08 0.09 0.08 0.09 0.09 0.21 0.34 0.34 0.36 0.36 0.35 5 B. tabaci A CT 1 0.00 0.01 0.01 0.02 0.00 0.22 0.34 0.35 0.36 0.35 0.38 6 B. tabaci A CR 0.00 0.01 0.02 0.01 0.23 0.35 0.36 0.37 0.36 0.39 7 B. tabaci A AZ 0.00 0.02 0.01 0.22 0.34 0.36 0.37 0.36 0.38 8 B. tabaci A PR 0.00 0.02 0.22 0.35 0.36 0.36 0.36 0.38 9 B. tabaci CT 2 0.00 0.22 0.34 0.35 0.36 0.36 0.38 10 B. tuberculata CT 0.00 0.28 0.28 0.37 0.36 0.38 11 A. socialis Mon 0.00 0.07 0.40 0.39 0.43 12 A. socialis CT 0.00 0.41 0.40 0.41 13 T. vaporariorum CT 0.00 0.00 0.39 14 T. vaporariorum AZ 0.00 0.38 15 T. variabilis CT 0.00 a. B, Bemisia; A, Aleurotrachelus; T, Trialeurodes.

254 Biotypes of Bemisia tabaci

Bemisia tuberculata 92 B. tabaci A 100 65 B. tabaci B

99 Aleurotrachelus socialis

Trialeurodes vaporariorum 100 90

T. variabilis

Magicicada septendecula

Stenocladius sp.

Figure 1. Cladogram showing the relationship between the whiteflies in this study. The cladogram is based on the most parsimonious tree inferred from the analysis of 485 base sites of a region of the mitochondrial 16S gene. Numbers above the branches indicate the level of statistical support for the corresponding node from 10,000 bootstrap replicates.

B were always grouped together. expanding base of phylogenetic B. tuberculata was grouped with the information on an increasing number of B. tabaci 92% of the time and gene sequences that exhibit different T. vaporariorum and T. variabilis were rates of change. grouped together 86% of the time. Using parsimony analysis, the This study generated additional relationship between the three genera evidence that the B. tabaci biotype B is of Bemisia, Trialeurodes and highly conserved throughout the Aleurotrachelus was not clear. The Americas. This was expected because it results of the distance analysis is a recent introduction. The (Table 1) show that Aleurotrachelus populations of B. tabaci biotype A are was closer to the genus Bemisia than more conserved than we would have to Trialeurodes. An unexpected result predicted from the studies that used was the relatively large mean esterase isozyme pattern as the means distance between T. vaporariorum and to detect diversity. Because the T. variabilis. expression of esterases can be induced by insecticide applications, their Gene sequences and phenotypes are probably not reliable for phylogenetic studies determining biotypes. Comparing the The use of DNA sequences has results of the molecular and esterase become increasingly more important analyses, many esterase isozyme as a tool to study the evolution, patterns are associated with the populations and systematics of B. tabaci biotype A based on DNA insects. Since the database for markers. sequence information is expanding exponentially, it is certain that these Even though, in the phylogenetic methods will be even more important analysis, T. vaporariorum and in the future. The several advantages T. variabilis were in the same clade, the to using DNA include the fact that absolute distance was fairly high. Since the genotype, and not the phenotype, the distance is equally great between is examined directly. There is also an T. vaporariorum, T. variabilis and the

255 Whiteflies and Whitefly-borne Viruses in the Tropics whiteflies in the other genera, there is a Analysis of molecular markers question of how closely related the to identify whiteflies in Latin members of Trialeurodes may be. America Further studies should determine if Four oligonucleotide primers (De Barro these two species are members of the and Driver, 1997) were tested for their same genera, or whether they should be utility in distinguishing B. tabaci placed into separate genera. biotype A and biotype B, and T. vaporariorum. In the analysis of PCR The application of RAPD for products using the primer H9 (Operon, identification of whiteflies USA), there were differences for the Whenever an increase in whiteflies range of whitefly species tested occurs or they begin to affect additional (Figure 2). For B. tabaci biotype B and crops, the introduction of the B. tabaci T. vaporariorum, there are prominent B biotype is suspected. Often, this is PCR products ca. 600 and 800 bp that the case but whitefly populations are can sometimes make distinguishing the affected by many environmental two species difficult. The unique PCR factors, including the cropping system product in B. tabaci biotype B ca. and the varieties grown. The molecular 950 bp and T. vaporariorum’s unique methods that generate DNA sequence PCR product at ca. 500 bp are data are time and labour intensive and important for distinguishing between are not suited to large scale monitoring these two species. Although confusion of populations. Rapid and reliable can occur in the interpretation between methods are needed to distinguish some species, the H9 primer was most between the most common whiteflies. useful in distinguishing between In Latin America, these are B. tabaci B. tabaci biotypes A and B. At 600 bp, biotypes A and B, and T. vaporariorum some PCR products are similar in size on most crops. in both biotypes but the biotype A has several unique PCR products, including Using RAPDs has several doublet bands at 300-350 bp. The advantages. Foreknowledge about any particular gene in the target organism is not needed. More than one primer can be used to increase confidence in MBABB the reliability of the method. With whiteflies and many other types of samples, the ability to store them for many months at room temperature in 70% ethanol facilitates shipping them across international borders and allows the analysis of large numbers of samples to be processed in an orderly manner.

RAPD was used (De Barro and Driver, 1997) to distinguish indigenous Australian populations of B. tabaci from the introduced B biotype. The authors reported on four oligonucleotide Figure 2. Using Operon primer H9, these are primers that they considered the most the RAPD-PCR DNA products from individual whiteflies. M:1kb markers useful for identifying the native B. (BRL), BA: Bemisia tabaci biotype A, tabaci from the B biotype. BB: Bemisia tabaci biotype B.

256 Biotypes of Bemisia tabaci biotype B has unique PCR products at This survey was extensive but not ca. 600, 700 and 900 bp compared exhaustive, therefore results should be with one product of ca. 850 bp in the interpreted as a representation of the biotype A. predominate whiteflies populations in various regions of Cuba, the Dominican The primer H16 (Figure 2) was Republic, Guatemala, Honduras, Costa most useful in distinguishing between Rica, Panama, Colombia and Venezuela. the whitefly species. While there can be some common bands in the 500 to Dominican Republic 1000 bp range for both B. tabaci In the departments of Azua, Barahona, biotypes, three products in B. tabaci Peravia, Santiago, Montecristo and San biotype B of ca. 350, 450 and 550 bp Juan, the predominant whitefly was were consistently useful for B. tabaci biotype B. Plants tested identification of the B biotype. include eggplant (Solanum melongena L.), tomato (Lycopersicon esculentum When primers F2 and F12 were Mill.), okra (Abelmoschus esculentus [L.] used, there were larger numbers of Moench), melon (Cucumis melo L.), PCR products. These can be used for cucumber (Cucumis sativus L. var. distinguishing the whitefly species but, sativus) and hibiscus (Hibiscus spp. L). because of the large numbers of bands, Only one sample of tomato in Azua was were generally less useful than H9 and classified a biotype A. In the lowland H16. When using RAPDs for the tropics of the Dominican Republic, the identification of the whiteflies in this biotype B was introduced nearly a study, analysing the individual decade ago and has nearly excluded the whiteflies with both the H9 and indigenous biotype A. In the department H16 primers is recommended. of Vega, common bean (Phaseolus vulgaris L.), tomato and potato (Solanum Mapping the Distribution tuberosum L.) were tested and T. vaporariorum was the only whitefly of Whitefly Species and found. Biotypes in Latin America

This survey was undertaken to map the Guatemala distribution and range of whiteflies in In the department of Zacapa, the various countries in Latin America. For predominant whitefly was B. tabaci. In this purpose, RAPDs were used to Zapcapa, the B biotype was present on characterize the principal whitefly pests all the hosts tested, which included in many countries of Central America okra, melon, watermelon (Citrullus and the Caribbean region. lanatus [Thunb.] Matsum. & Nakai) and cucumber. In Japala, both B. tabaci The survey using molecular biotype B and T. vaporariorum were markers complemented the activities found. The B biotype was found on where whitefly specimens were okra, melon, watermelon, cucumber, analysed by light microscopy. For the tomato, eggplant, tobacco (Nicotiana molecular analysis, at least one and tabacum L.) and weed species. often two primers were used and the T. vaporariorum was the predominate results were compared to the whitefly in several host plants including morphological identification. The use of paw paw (Carica papaya L.), “tomate- RAPD data was the only method to manzana” (Lycopersicon esculentum var. distinguish between the biotypes A and beef tomato), cherry tomato B in the Bemisia complex. (Lycopersicon esculentum var.

257 Whiteflies and Whitefly-borne Viruses in the Tropics cerasiforme [Dunal] A. Gray), common introduction of B. tabaci biotype B in bean and some samples of tomato. Panama.

Honduras Cuba In Honduras, the predominant whitefly B. tabaci biotype B was the only was B. tabaci biotype A. Most of the whitefly found in the samples from samples were common bean, tomato Cuba. The B biotype has been in Cuba and cucumber from the Comayagua for nearly 10 years and clearly has Department. The paw paw in become the predominant whitefly. Comayagua was host to T. vaporariorum. On chilli peppers Colombia (Capsicum spp. L.) in the Francisco Morazan Department, the A biotype Most of the samples tested were from was also the only whitefly identified. the north coastal region of Colombia, where the principal whitefly is B. tabaci Costa Rica biotype B. The B biotype is affecting tomato, col (Brassica oleracea L. var. In the lowland tropics of Costa Rica, capitata L.), eggplant, melon and B. tabaci biotype A is still the cassava (Manihot esculenta Crantz). The predominant whitefly. In the A biotype is still present but the departments of Guanacaste, San José, introduction of the B biotype is Heredia and Alajuela, most of the relatively recent. samples were biotype A. The A biotype was found on common bean, melon, Venezuela watermelon, tomato, chilli peppers, cucumber and others. In the Samples were collected from the departments of Arajuel and departments of Zamora and Jiménez, Puntarenas, the B biotype was present. where the principal whitefly was the It was the only whitefly found in the B biotype. In the Urdaneta two samples from Arajuel but in Department, there was a mixture of the Puntarenas both biotype A and B were A and B biotypes. Only a few samples present. T. vaporariorum was present in were analysed and more are needed the Alajuela Department and was before conclusions should be made. common in Cartago where it was found on chilli peppers and tomato. A Regional View of El Salvador Whitefly Populations Only a limited number of samples (15) Most of the samples that were have been tested and all were B. tabaci determined to be in the B. tabaci biotype B. complex by morphological methods were classified as either biotype A or Panama B (Table 2). In general, about 15% of The principal whitefly in the the samples could not be amplified; Department of Chiriqui was only 23% of the samples from T. vaporariorum. In the other regions of Venezuela could be amplified. This may the country, B. tabaci biotype B was demonstrate the importance of the the predominant whitefly. Populations proper sample handling. About 5% of of B. tabaci biotype A are still found the samples also had RAPD PCR and in some areas the populations are products that could not be identified. mixed. This suggests a relatively recent These may well be either different

258 Biotypes of Bemisia tabaci

Table 2. The species of whiteflya by country (%) as determined using random amplified polymorphic DNA (RAPD) analysis.

Country Number B. tabaci B. tabaci T. vaporariorum NIb NBc of samples biotype A biotype B

Guatemala 185 8.1 60.5 26.0 5.4 0 Cuba 44 0 100.0 0 0 0 Dominican Rep. 106 13.2 81.1 0 0 5.7 Colombia 173 0.6 63.6 14.4 5.8 15.6 Venezuela 262 4.6 18.3 0 0 77.1 Costa Rica 160 61.3 6.9 12.5 3.1 16.2 Panama 198 4.6 33.3 40.9 9.1 12.1 El Salvador 15 0 100.0 0 0 0 Honduras 138 88.4 2.9 0 7.3 1.4 a. Bemisia tabaci, Trialeurodes vaporariorum. b. Samples with amplified PCR products but unable to identify. c. No products were amplified. species or unique biotypes. This level of documents the expanding range of uncertainty did not detract from the B. tabaci biotype B (Polston and survey and those unidentified samples Anderson, 1997) and adds details to are a source of potentially different the distribution within the countries in whitefly populations that merit further the study. In Honduras and Costa study. Rica, the A biotype is still the major whitefly present. It is probably just a B. tabaci biotype A, B. tabaci matter of time until the B biotype biotype B and T. vaporariorum were the becomes the principal whitefly in those principal whiteflies found in the region. areas. Nevertheless, these are the Even at the department level, a zones that need to be monitored closely predominant whitefly usually could be and where strategies to decrease the identified. Occasionally, within the probability of introduction of the B same region, mixed populations were biotype should be developed. present but these were normally separated on different hosts. T. Developing a specific marker vaporariorum was present in many to identify whiteflies in the regions but tends to be at higher Bemisia complex elevations. Only in the mid-altitudes do B. tabaci and T. vaporariorum RAPDs are produced by using short populations overlap. In some countries, oligonucleotides, generally of 10 base including Cuba and the Dominican pairs. This allows the generation of Republic, the exclusion of the B. tabaci many amplified PCR products, using biotype A is almost complete, at least just one primer. The lack of specificity on the principal crop species tested for of the short oligonucleotides can lead this survey. The pattern in Central to results that are difficult to interpret America is more complex. In Cuba, because of similar size PCR products. Guatemala, Panama and the north cost Also, shorter oligonucleotides are even of Colombia, the process of domination more sensitive to single base changes. of the B biotype appears to be well RAPD markers can be converted in a advanced. In Colombia, biotype B is a sequence characterized amplified recent introduction (Quintero et al., region (SCAR). This specific marker 1998) and already is the predominant utilizes a pair of oligonucleotides whitefly in the region. This survey generally 20-25 base in length and

259 Whiteflies and Whitefly-borne Viruses in the Tropics amplifies a single region of a genome A and T. vaporariorum. Each set of (Ohmori et al., 1996). This makes primers will amplify bands of different interpretation of the results much sizes. This will allow the primers to be simpler than with RAPDs. Instead of a mixed together and will distinguish group of bands with a degree of between the whiteflies in a single PCR variability, SCAR produces single reaction. Thus more extensive surveys bands in the target organism. possible will be possible and the technique will be more consistent in all The SCAR for the B. tabaci biotype the laboratories that need the B was developed by cloning the RAPD diagnostic capability to rapidly identify PCR products of the primer H9. The whitefly pests. cDNA clones were analysed and sequenced. The primers were designed and tested. One set of primers only Conclusions amplified B. tabaci biotype B (Figure 3). This set of primers did not amplify any Our detailed molecular analysis is tested population of B. tabaci biotype A helping define both the range and the or T. vaporariorum. Further testing is variability of B. tabaci biotypes A and B needed to confirm that these primers in Latin America. Similar studies in amplify this PCR product in other regions of the world have been geographically separated populations of made and will allow a global view of the B. tabaci biotype B. Given the lack of Bemisia complex. The molecular diversity in the evolutionary studies, method of RAPDs proved essential in this SCAR is expected to be useful for distinguishing between the B. tabaci rapidly identifying all populations of biotype A and B but, because of the biotype B. nature of the method, comparing results between laboratories is difficult. The current research is directed to Therefore we are developing SCARs to developing SCARs for B. tabaci biotype simplify the identification of whiteflies.

M123456M M1 2 3 4 5M

Figure 3. Two SCARs were developed to distinguish Bemisia tabaci biotype A (1) and Bemisia tabaci biotype B (2). Lane 1 Bemisia tabaci biotype A, lane 2 Bemisia tabaci biotype B, lane 3 Bemisia tuberculata, lane 4 Trialeurodes vaporariorum, lane 5 Aleurotrachelus socialis, 6:DNA free reaction and M:1kb markers (BRL).

260 Biotypes of Bemisia tabaci

Already we can identify B. tabaci De Barro, P. J.; Driver, F.; Trueman, J. W. biotype B using a SCAR, and we are H.; Curran, J. 2000. Phylogenetic making progress in the development of relationships of world populations of Bemisia tabaci (Gennadius) using SCARs for B. tabaci biotype A and ribosomal ITS1. Molec. Phylo. Evol. T. vaporarorium. Specific unambiguous 16:29-36. results of one amplified band for each species or biotype of whitefly will Felsenstein, J. 1993. PHYLIP (Phylogeny simplify the interpretation of results Inference Package) version 3.5c. and become a rapid and relatively Distributed by author. Department inexpensive method to use in surveys of Genetics, University of Washington, Seattle, USA. of whiteflies. Frohlich, D. R.; Torres-Jerez.; Bedford, I. D.; Markham, P. G.; Brown, J. K. References 1999. A phylogeographic analysis of the Bemisia tabaci species complex Brown, J. K.; Coats, S.; Bedford, I. D.; based on mitochondrial DNA Markham, P. G.; Bird, J.; Frohlich, markers. Mol. Ecol. 8:1683-1691. D. R. 1995. Characterization and distribution of esterase Gawell, N. J.; Bartlett, A. C. 1993. electromorphs in the whitefly, Characterization of differences Bemisia tabaci (Genn.)(Homoptera: between whiteflies using RAPD Aleyrodidae). Biochem. Genet. PCR. Insect Mol. Biol. 2:33-38. 33:205-214. Gilbertson, R. L.; Rojas, M. R.; Russell, D. Calvert, L. A.; Cuervo, M.; Arroyave, J. A.; R.; Maxwell, D. P. 1991. Use of the Constantino, L. M.; Bellotti, A.; asymmetric polymerase chain Frohlich, D. 2001. Morphological reaction and DNA sequencing to and mitochondrial DNA marker determine genetic variability of bean analyses of whiteflies (Homoptera: golden mosaic geminivirus in the Aleyrodidae) colonizing cassava and Dominican Republic. J. Gen. Virol. beans in Colombia. Ann. Entomol. 72: 2843-2848. Soc. Am. 94(4):512-519. Mohanty, A .K.; Basu, A. N. 1986. Effect Campbell, B. C.; Steffen-Campbell, J. D.; of host plants and seasonal factors Gill, R. J. 1994. Evolutionary origin on intraspecific variations in pupa of whiteflies (Hemiptera: morphological of the whitefly vector, Sternorryncha: Aleyrodidae) Bemisia tabaci (Genn.) (Homoptera: inferred from 18S rDNA sequences. Aleyrodidae). J. Ent. Res. 10:19-26. Insect Mol. Biol. 3:73-89. Mound, L. A. 1963. Host-correlated David, B. V.; Ananthakrishnan T. N. variation in Bemisia tabaci (Genn.) 1976. Host correlated variation in (Homoptera Aleyrodidae). Proc. R. Trialeurodes rara (Singh) and Ent. Soc. London. 38:171-180. Bemisia tabaci (Gennadius) (Homoptera: Aleyrodidae). Curr. Sc. Ohmori, T.; Murata, M.; Motoyoshi, F. (Bangalove) 45:223-225. 1996. Molecular characterization of random amplified polymorphic DNA De Barro, P. J.; Driver, F. 1997. Use of (RAPD) and sequence characterized random amplified polymorphic DNA amplified region molecular (SCAR) polymerase chain reaction (RAPD markers linked to the Tm-1 locus in PCR) to distinguish the B biotype tomato. Theor. Appl. Genet. 92: from other biotypes of Bemisia 151-156. tabaci (Gennadius) (Hemiptera: Aleyrodidae). Aust. J. Entomol. 36:149-152.

261 Whiteflies and Whitefly-borne Viruses in the Tropics

Perring, T. M.; Cooper, A. D.; Rodríguez, Swofford, D. L. 1999. PAUP*. Phylogenetic R. J.; Farrar, C. A.; Bellows, T. S. Analysis Using Parsimony (*and 1993. Identification of a whitefly other methods). Version 4. Sinauer species by genomic and behavioural Associates, Sunderland, MA, USA. studies. Science 259:74-77. Thompson, J. D.; Higgins, D. G.; Gibson, Polston, J. E.; Anderson, P. K. 1997. The T. L. 1994. CLUSTAL W: Improving emergence of whitefly-transmitted the sensitivity of progressive geminiviruses in tomato in the multiple sequence alignment western hemisphere. Plant Dis. through sequence weighting, 81:1358-1369. positions-specific gap penalties and weight matrix choice. Nucleic Acids Quintero, C.; Cardona, C.; Ramírez, D.; Res. 22:4673-4680. Jiménez, N. 1998. Primer registro del biotipo B de Bemisia tabaci Wool, L.; Calvert, L.; Constantino, L. M.; (Homoptera: Aleyrodidae) en Bellotti, A. C.; Gerling, D. 1994. Colombia. Rev. Colomb. Entomol. Differentiation of Bemisia tabaci 24:23-28. (Genn.) populations in Colombia. J. Appl. Entomol. 117:122:134. Rosell, R. C.; Bedford, I. D.; Frohlich, D. R.; Gill, R. J.; Brown, J. K.; Xiong, B.; Kocher, T. D. 1991. Markham, P. G. 1997. Analysis of Comparison of mitochondrial DNA morphological variation in distinct sequences of seven morphospecies populations of Bemisia tabaci of black flies (Diptera: Simuliidae). (Homoptera: Aleyrodidae). Ann. Genome 34:306-311. Entomol. Soc. Am. 90:575-589.

Sanger, F.; Nicklen, S.; Coulson, A. R. 1977. DNA sequencing with chain- terminating inhibitors. Proc. Natl. Acad. Sci. USA. 74:5463-5467.

262 Conclusions and Recommendations

CHAPTER 3.14 Conclusions and Recommendations

Francisco Morales*

Despite over 40 years of research on (Cucumis melo L.), common bean, Bemisia tabaci (Gennadius) tomato, eggplant (Solanum melongena (Homoptera:Aleyrodidae) as a pest and L.), cucurbits and sweet and hot vector of plant viruses, this whitefly peppers (Capsicum spp. L.) has been species continues to cause reduced greatly or completely considerable damage to food and abandoned during this period of the industrial crops in Mexico, Central year because of the frequent outbreaks America and the Caribbean. B. tabaci of whiteflies and whitefly-borne viruses. first became a pest of cotton The development of B. tabaci from a (Gossypium hirsutum L.) in Middle harmless insect to a major pest is America, and then a major vector of closely related to the introduction and plant viruses in common bean intensive use of pesticides after the (Phaseolus vulgaris L.) and tomato Second World War. Pesticide abuse (Lycopersicon esculentum Mill.) remains a major factor in B. tabaci and plantings throughout the region. The begomovirus epidemics in this region. expansion of non-traditional crops in Whiteflies are known to develop the 1980s, mainly vegetables for resistance to insecticides, the frequent export, further aggravated the problem use of which eliminates natural by providing more reproductive hosts predators of B. tabaci. It is not for the whitefly pest and a higher uncommon to see farmers spraying demand for pesticides. The insecticides on susceptible crops every introduction in the early 1990s of a other day and even daily, up until more aggressive and polyphagous harvest time, in open defiance of food biotype (B) of B. tabaci in the Americas safety regulations. The high levels of further jeopardized past efforts to pesticide residues in tomato, pepper, control this pest. common bean, eggplant and other vegetables attacked by B. tabaci In this region, whitefly-borne disqualifies these products for sale in geminiviruses (begomoviruses) are international markets and constitutes a more prevalent during the dry months serious health hazard for the millions of the year (November through April), of rural and urban consumers in the when B. tabaci populations encounter region, where produce is not tested for more favourable weather. The area toxic residues. planted to crops such as melon But the main reason for the persistence of these pest problems is * Centro Internacional de Agricultura Tropical the lack of qualified technical (CIAT), Cali, Colombia. assistance for farmers, particularly

263 Whiteflies and Whitefly-borne Viruses in the Tropics those who have attempted to diversify due to various factors. The constant their subsistence crops with more flow of improved cultivars developed in profitable cash crops such as tomato the past through conventional plant and pepper. This situation is a breeding methods has been greatly reflection of the increasingly reduced because of the radical changes deteriorating economic situation of the that have occurred in the past decade region since the mid 1970s, which has in favour of molecular breeding negatively impacted on the capacity of methods. Second, technical assistance national agricultural research to small-scale farmers is practically nil, institutions and international which leaves plant protection in the agricultural research centres to hands of pesticide salespeople. conduct research and provide the Susceptible crop-free periods reduce necessary technical assistance to whitefly incidence but do not increase farmers affected by severe the productivity of prime agricultural phytosanitary problems such as land that must be devoted to less whiteflies and begomoviruses. profitable crops or remain idle during Moreover, the diversion of foreign aid several months of the year. The new funds from crop improvement to insecticides in the market are effective natural resource management has for whitefly control but are very further contributed to increased yield expensive for resource-poor farmers. losses due to pest problems, abandonment of profitable crops grown The Tropical Whitefly Integrated by small-scale farmers, higher Pest Management (TWF-IPM) Project production costs and considerably has successfully identified, tested and more environmental and food validated the most efficient and contamination. sustainable whitefly control methods used in over 10 countries in Middle Whitefly and begomovirus control America. So far, the most valuable IPM strategies in the region differ according measure identified is genetic resistance to countries, agricultural areas and to control whitefly-transmitted viruses, crops. In north-western Mexico, the particularly in the case of common use of virus-resistant common bean bean. Unfortunately, practically no varieties and more efficient insecticides crop improvement work is being done for vegetables has greatly reduced the in the case of vegetables such as impact of these pests. In the Yucatán tomato and pepper despite their Peninsula, Mexico, small-scale farmers tropical American origin. Most were using physical control methods vegetable varieties are bred in during the first 3 weeks after temperate countries and thus are not transplant of tomato and pepper. In the adapted to the tropical conditions and Dominican Republic, a crop-free period pests that exist in the region. The TWF- was legally imposed throughout the IPM Project has successfully used country to break the cycle of the physical barriers (micro-tunnels) to whitefly before the main planting protect sensitive crops during the first season. And in most Central American critical month of their life cycle. This countries, susceptible crops were only strategy also protects susceptible planted during the rainy season, when plants against other insect-borne whitefly populations are at their lowest viruses and pests, and greatly level. These whitefly/virus control increases their productivity per unit strategies have reduced, but not area with minimum inputs. The eliminated, the production losses identification and use of bio-control associated with outbreaks of B. tabaci agents of B. tabaci has been advanced

264 Conclusions and Recommendations greatly in the region, particularly in systems with more profitable crops in countries such as Cuba and the order to maximize the income derived Dominican Republic. Finally, an from their limited land resources. enhanced understanding of the whitefly Traditional crops are the basis of food species, biotypes and plant viruses security but they are merely present in the region, has helped subsistence crops. High-value crops, understand the ecology and particularly vegetable and fruit crops, epidemiology of these crop production can help small-scale farmers improve problems and, more important, has their livelihoods but this can only be contributed to the implementation of achieved with proper and timely sustainable and efficient IPM methods. technical assistance provided by multi- disciplinary groups of agricultural It is imperative to re-orient specialists. Finally, there is a vacuum agricultural research in the region by in the area of agricultural economics understanding that spending our and marketing for small-scale farmers. limited resources on purely Unless these farmers are incorporated environmental or sociological issues, into the market chains, they will be at without a food production component, the mercy of the vagaries of the open only perpetuates misery and increases markets and intermediaries who profit environmental degradation and human from the misery and lack of market health problems. There is considerable knowledge of the resource-poor farmer. need to strengthen national and The TWF-IPM Project expects to international research and extension address these issues and scale up the services to provide timely and effective dissemination of the IPM measures technical assistance to small-scale validated to date in Central America, farmers. We must help resource-poor Mexico and the Caribbean region. farmers to diversify their cropping

265 BLANCA 266 SECTION FOUR

Whiteflies as Pests of Annual Crops in the Tropical Highlands of Latin America BLANCA 268 Introduction

CHAPTER 4.1 Introduction

César Cardona*

According to Strong et al. (1984), B. tabaci and T. vaporariorum as virus 724 species of whiteflies (Homoptera: vectors, emphasis is on the importance Aleyrodidae) have been described from of these insects as direct pests causing the tropics but only 420 from mechanical damage. The whiteflies temperate regions. Most of these affecting cassava (Manihot esculenta insects do not represent an economic Crantz), a semi-perennial crop, are not threat to agriculture and only a discussed. handful can be described as serious pests. In a recent survey conducted in Central America and Colombia, Legumes and Caballero (1992) listed the Horticultural Crops in the 30 commonest and economically most Andean Highlands important whitefly species recorded from 84 host plant species. Among The Andes encompass a wide range of these, the sweetpotato whitefly Bemisia latitudes and climates but in the tabaci (Gennadius), the greenhouse tropics especially, altitude is the main whitefly Trialeurodes vaporariorum determinant of temperature and (Westwood) and the citrus whitefly biological character. Areas between Aleurocanthus woglumi Ashby were 1000 and 2000 m above sea level are regarded as key pests. Of lesser known as highlands. Those above importance were Trialeurodes 2000 m have a quasi-temperate climate abutiloneus (Haldeman) and three and are collectively known as the “high species affecting cassava: Andes”. Transitions from tropical Aleurotrachelus socialis Bondar, lowlands to temperate highlands may Bemisia tuberculata Bondar and be abrupt and the alpine zones form Trialeurodes variabilis (Quaintance). ecological islands, large or small. As indicated by Winograd et al. (1998), the T. vaporariorum and B. tabaci are Andean backbone and its lateral ranges by far the most important whitefly contribute remarkable diversity to the species affecting annual crops in the ecosystems of Colombia and Ecuador. Latin American highlands and these In parts of Colombia, small ridges and two species are the focus of the present spurs connect the three main ranges, review. While reference is made to circumscribing a series of intermountain basins, each of which hosts a distinctive local community of natural biodiversity. Agricultural * Centro Internacional de Agricultura Tropical systems and their ecology are (CIAT), Cali, Colombia. correspondingly diverse.

269 Whiteflies and Whitefly-borne Viruses in the Tropics

The tropical highlands of Colombia 1998; Salazar et al., 2000) but it is and Ecuador include extensive more important as a direct pest of agricultural areas with mean several crops. This whitefly was viewed temperatures of 12-20 °C, variable as a minor pest until the early 1980s rainfall patterns (700-1600 mm) and (Cardona, 1995). Since then, relative humidity ranging from 50% to T. vaporariorum has become a key pest 85%. Small-scale farmers grow on dry beans, snap bean, tomato, common bean and snap bean potato and ornamentals and is (Phaseolus vulgaris L.), tomato considered the most important whitefly (Lycopersicon esculentum Mill.), potato species in the tropical highlands of (Solanum tuberosum L.) and other South America. Damage is due to the horticultural crops, both for home continuous sucking of sap from the consumption and for sale. Crops are phloem by nymphs and adults and to usually grown on steep, erosion-prone the abundant excretion of honeydew slopes, with no irrigation and little or that falls on leaves and fruits and no use of fertilizers. serves as a substrate for fungi that block photosynthesis. Both the yield In contrast, the lowlands of and quality of crops are affected. A Colombia and Ecuador experience series of trials in Colombia (Cardona et mean temperatures of 26-28 °C, al., 1993; Prada et al., 1993) showed similarly variable rainfall patterns that high infestation levels (up to (600-1600 mm) and prolonged dry 450 nymphs/leaf) can reduce the yields seasons that favour the buildup of of common bean by 38% and of snap whitefly populations. Cotton bean by 54%. (Gossypium hirsutum L.), soybean (Glycine max [L.] Merr), melon (Cucumis Whitefly problems in snap bean melo L.) and, to a lesser extent, tomato production in Colombia have increased are grown on large farms that occupy a dramatically since the early 1980s and significant proportion of the illustrate the need for integrated pest agricultural land. Many small-scale management (IPM) work on whiteflies. farmers also grow less than 1 hectare In an attempt to reduce pesticide use of legumes, vegetables or tobacco in whitefly-affected areas, a pilot (Nicotiana tabacum L.). project, funded by the International Development Research Centre (IDRC) of Canada, was initiated in the Status of Research on Sumapaz region of central Colombia. Trialeurodes and Bemisia Surveys showed that all of 286 farmers interviewed were using insecticides as The centre of origin for T. vaporariorum the sole method to manage has not been identified. According to whiteflies. Insecticide use averaged Vet et al. (1980), the species is most 11 applications in a 70-day cropping probably indigenous to tropical and cycle. Most of the insecticides used subtropical America. T. vaporariorum, a were organophosphates and well-known pest of greenhouse crops in carbamates belonging to toxicological Europe, has long been recorded on category I (highly toxic) as defined by annual field crops in highland areas of Metcalf (1994): those insecticides with

Colombia (Posada, 1976) and Ecuador acute oral LD50 values of less than

(Merino and Vásquez, 1962). It is 50 mg/kg or dermal LD50 values of less confirmed as a competent vector of than 200 mg/kg. Field tests indicated numerous plant viruses (Díaz et al., that seven of the 10 most widely used 1990; Duffus, 1987; 1996; Wisler et al., insecticides were ineffective for whitefly

270 Introduction control, due to resistance (Cardona, status was attributed to the 1995). Research showed that simple introduction into the region of the B sanitation measures such as leaf biotype (Quijije et al., 1995), which was roguing and destruction of crop first detected in Colombia in 1997 residues reduced overall whitefly (Quintero et al., 1998). infestation levels if implemented on a community-wide basis. In addition, selective use of granular insecticides A New Foundation for delayed the need for the first foliar Whitefly IPM in the Andes insecticide application, enhancing natural enemy activity. A relatively In view of the increasing importance of simple IPM package based on cultural whiteflies in the Andean highlands, the control and sanitation practices, timely launching in 1997 of a new project for application of effective insecticides and “sustainable integrated management of reliance on natural biological control whiteflies as pests and vectors of plant resulted in a 66% reduction in viruses in the tropics” by the insecticide use (Cardona, 1995). Systemwide Programme on Integrated Farmers’ participation in the overall Pest Management was especially process was essential for widespread timely. A sub-project was established, adoption of the system proposed. under the co-ordination of the Centro However, it was concluded that, before Internacional de Agricultura Tropical continuing with IPM training and (CIAT), to look specifically at “Whiteflies scaling up, it was necessary to verify as pests in the tropical highlands of that patterns of T. vaporariorum Latin America”. distribution, reproductive hosts plants, perceptions of the problem and The purpose of the diagnostic insecticide use were similar throughout phase of this sub-project (and of others the Andean region. addressing different whitefly problems around the world) was to provide a B. tabaci also has been sound basis for future IPM efforts by documented as a pest of several crops gathering and analysing baseline data in Colombia (Posada, 1976; Bolaño, and so being able to characterize 1997) and Ecuador (Merino and properly the nature of major whitefly Vásquez, 1962). Except for sporadic problems. Specifically, in the Andean outbreaks on cotton in Colombia highlands, there was a need to describe (Alcaraz et al., 1990), this insect was the patterns of distribution of generally regarded as a secondary pest T. vaporariorum and B. tabaci, until 1993 when serious outbreaks especially in relation to altitude, as well occurred in the western provinces of as the distribution of the newly Manabí, Guayas, and Los Ríos in introduced B biotype of B. tabaci. Ecuador. Mendoza et al. (1995) Information on the range of estimated that vegetable growers in the reproductive host plants and Guayas region lost US$400,000 per identification of natural enemies was year as a result of whitefly attacks. Up also desired. Additional socio-economic to 10,000 ha of soybean were destroyed information on crop losses, farmers’ in the Guayas and Los Ríos provinces perception of the problems and (Mendoza, 1996). Similar outbreaks patterns of insecticide use, as well as occurred in 1996 in the north-western biological assessment of levels of Departments of Sucre, Córdoba and insecticide resistance, were needed as a Atlántico in Colombia (Quintero et al., foundation for moving forward to more 2001). The change in B. tabaci’s pest widespread adoption of IPM.

271 Whiteflies and Whitefly-borne Viruses in the Tropics

Surveys were conducted between Cardona, C.; Rodríguez, A.; Prada, P. C. October 1997 and December 1998 in 1993. Umbral de acción para el whitefly-affected areas of Colombia and control de la mosca blanca de los invernaderos, Trialeurodes Ecuador while insecticide resistance vaporariorum (Westwood) testing continued through May 1999. (Homoptera: Aleyrodidae), en The research was carried out in close habichuela. Rev. Colomb. Entomol. collaboration among staff of CIAT and 19(1):27-33. partner organizations, the Instituto Nacional Autónomo de Investigaciones Díaz, M. C.; Pulgarín, J. M.; Saldarriaga, Agropecuarias (INIAP) in Ecuador and A. V. 1990. Relaciones insecto- patógeno en el problema del the Corporación Colombiana de amarillamiento de las venas de la Investigación Agropecuaria papa. Rev. Colomb. Entomol. (CORPOICA) in Colombia. The methods 16(2):3-14. used and the results of the work are reported in the following two chapters. Duffus, J. E. 1987. Whitefly transmission of plant viruses. Curr. Topics Vector Res. 4:73-94.

References Duffus, J. E. 1996. Whitefly-borne viruses. In: Gerling, D.; Mayer, R.T. Alcaraz, H.; Cardona, C.; Revelo, F.; (eds.). Bemisia: 1995. Taxonomy, Revelo, R.; Herrera, M.; Alvarez, A.; biology, damage, control and Siabato. A. 1990. Plagas management. Intercept, Andover, secundarias o de incidencia GB. p. 252-263. ocasional. In: Bases técnicas para el cultivo del algodón en Colombia. Mendoza, J. 1996. Qué está pasando con Cuarta edición. Federación la mosca blanca en el Ecuador? Nacional de Algodoneros, Bogotá, Rev. INIAP 8:8-10. CO. p. 519-524. Mendoza, J.; Valarezo, O.; de López, M. Bolaño, R. E. 1997. Determinación de A.; Quijije, R.; Cañarte, E.; Alcaraz, niveles de daño económico de V. 1995. Reporte de Ecuador. Bemisia tabaci en tomate en el norte Memoria IV Taller Latinoamericano del Cesar, Colombia. MIP 46:26-33. sobre moscas blancas y geminivirus, 16-18 octubre 1995, Caballero, R. 1992. Whiteflies Zamorano, HN. CEIBA 36(1):13-15. (Homoptera: Aleyrodidae) from Central America and Colombia Merino, G.; Vásquez, V. 1962. including slide-mounted pupal and Identificación de algunas de las field keys for identification, field nuevas especies de insectos characteristics, hosts, distribution, coleccionadas en Ecuador. Bol. natural enemies, and economic Técnico no. 7, Dirección General de importance. M.Sc. thesis, Kansas Agricultura y Servicio Cooperativo State University, USA. 201 p. Interamericano de Agricultura, Quito, EC. 35 p. Cardona, C. 1995. Manejo de Trialeurodes vaporariorum (Westwood) en frijol en Metcalf, R. L. 1994. Insecticides in pest la zona Andina: Aspectos técnicos, management. In: Metcalf, R. L.; actitudes del agricultor y Luckmann, W. H. (eds.). transferencia de tecnología. Introduction to insect pest Memoria IV Taller Latinoamericano management. 3rd edition. John sobre moscas blancas y Wiley, NY, USA. p. 245-284. geminivirus, 16-18 octubre 1995, Zamorano, HN. CEIBA 36(1):53-64.

272 Introduction

Posada, L. 1976. Lista de insectos Salazar, L. F.; Muller, G.; Querci, M.; dañinos y otras plagas en Colombia. Zapata, J. L.; Owens, R. A. 2000. Bol. Técnico no. 43, Instituto Potato yellow vein virus: Its host Colombiano Agropecuario (ICA). range, distribution in South Bogotá, CO. 662 p. America and identification as a crinivirus by Trialeurodes Prada, P. C.; Rodríguez, A.; Cardona, C. vaporariorum. Ann. Appl. Biol. 1993. Evaluación de un sistema de 137:7-19. manejo integrado de la habichuela en la provincia de Sumapaz Strong, D.; Lawton, J.; Southwood, R. (Cundinamarca). Rev. Colomb. 1984. Insects on plants: Entomol. 19(2):58-63. Community patterns and mechanisms. Blackwell Scientific, Quijije, R.; Mendoza, J.; Gómez, A. 1995. Oxford, GB. Ciclo biológico de Bemisia argentifolii en condiciones de Vet, L. E.; van Lenteren, J. C.; Woets J. laboratorio. Memoria IV Taller 1980. The parasite-host Latinoamericano sobre moscas relationship between Encarsia blancas y geminivirus, 16-18 formosa (Hymenoptera: Aphelinidae) octubre 1995, Zamorano, HN. and Trialeurodes vaporariorum CEIBA 36(1):84. (Homoptera: Aleyrodidae). Z. Angew. Ent. 90:26-51. Quintero, C.; Cardona, C.; Ramírez, D.; Jiménez, N. 1998. Primer registro Winograd, M.; Farrow, A.; Eade, J. 1998. del biotipo B de Bemisia tabaci Atlas de indicadores ambientales y (Homoptera: Aleyrodidae) en de sustentabilidad para América Colombia. Rev. Colomb. Entomol. Latina y el Caribe. Versión 1. 24(1-2):23-28. Centro Internacional de Agricultura Tropical (CIAT)-Progama de las Quintero, C.; Rendón, F.; García, J.; Naciones Unidas para el Medio Cardona, C.; López-Avila, A.; Ambiente (PNUMA), Cali, CO. Hernández, P. 2001. Especies y 1 CD-ROM. biotipos de moscas blancas (Homoptera: Aleyrodidae) en Wisler, G.; Duffus, J. E.; Liu, H-Y.; Li, R. cultivos semestrales de Colombia y H. 1998. Ecology and epidemiology Ecuador. Rev. Colomb. Entomol. of whitefly-transmitted 27(1-2):27-31. closteroviruses. Plant Dis. 82: 270-275.

273 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 4.2 Colombia and Ecuador

César Cardona*, Aristóbulo López-Avila** and Oswaldo Valarezo***

As part of the diagnostic phase of the The surveys focused on the tropical Tropical Whitefly Integrated Pest highlands (1000-3000 m above sea Management (TWF-IPM) Project (see level) along the Andean corridor of Introduction and Chapter 4.1, this Colombia and Ecuador (0° 30'-11°N, volume), partners in the sub-project 73°-78° 30'W). Within this zone, concerned with “Whiteflies as pests in surveys covered areas representing the tropical highlands of Latin America” about 97% of whitefly-affected areas conducted extensive field surveys in (Figure 1). Representative sampling Colombia and Ecuador from October also was carried out in the lowlands 1997 to December 1998. The sub- and mid-altitudes in order to construct project was co-ordinated by the Centro an altitudinal gradient from sea level Internacional de Agricultura Tropical up to 3000 m. Samples were collected (CIAT) and carried out in collaboration in mid-altitude valleys within the with the Ecuadorian national research highland tropics of Colombia organization: the Instituto Nacional (400-1000 m above sea level). And, in Autónomo de Investigaciónes the tropical lowlands (less than 400 m Agropecuarias (INIAP), and the above sea level), surveys were Colombian national research conducted along the northern (Atlantic) organization: the Corporación coast of Colombia (7° 30'-11°N, Colombiana de Investigación 72°-77°W) and the Pacific coast of Agropecuaria (CORPOICA). The present Ecuador (0°-1° 30'S, 79°-80° 30'W), chapter summarizes the diagnosis and including the Galápagos Islands of characterization of problems caused by Ecuador (2°-3°S, 90°-90° 30'W) in the the greenhouse whitefly Trialeurodes Pacific Ocean. INIAP staff conducted vaporariorum (Westwood) as a pest of field surveys and interviews with annual field crops in Colombia and farmers along the coast of Ecuador. Ecuador, and its relationship to the CIAT personnel sampled the northern sweetpotato whitefly Bemisia tabaci highlands of Ecuador. CORPOICA and (Gennadius). CIAT staff conducted fieldwork in Colombia.

* Centro Internacional de Agricultura Tropical Surveys followed the standardized (CIAT), Cali, Colombia. ** Corporación Colombiana de Investigación methodology agreed among the project Agropecuaria (CORPOICA), Bogotá, partners. Twenty survey areas were Colombia. defined in Colombia and Ecuador *** Instituto Nacional Autónomo de Investigaciones Agropecuarias (INIAP), based on the importance of annual Manabí, Ecuador. crops and previous knowledge of

274 Colombia and Ecuador

Caribbean Sea B biotype B. tabaci N A and B biotypes B. tabaci and T. vaporariorum B biotype B. tabaci and T. vaporariorum T. vaporariorum

A and B biotypes B. tabaci

Pacific Ocean Colombia

Ecuador

Figure 1. Survey area and distribution of whitefly species (Bemisia tabaci and Trialeurodes vaporariorum) in Colombia and Ecuador. Uncoloured areas were not sampled. whitefly incidence. Nine areas were agricultural areas along the three chosen to represent the lowlands along mountain ranges in Colombia. the Pacific coast of Ecuador and the Highland areas in both countries are Atlantic coast of Colombia where cotton occupied by small-scale farmers who (Gossypium hirsutum L.), soybean grow common and snap bean (Glycine max [L.] Merr.) and tomato (Phaseolus vulgaris L.), tomato, potato (Lycopersicon esculentum Mill.) are (Solanum tuberosum L.) and other grown on large farms that occupy a horticultural crops for consumption significant proportion of the and sale. Between 10 and 20 data sets agricultural land. Eleven areas were were collected and processed in each chosen to represent the highlands of survey area. A complete data set northern Ecuador and the main consisted of a producer questionnaire

275 Whiteflies and Whitefly-borne Viruses in the Tropics and biological samples of whiteflies and niches and habitats (Figure 2). In the natural enemies from one of that tropical lowlands, 91.5% of samples producer’s fields. Three hundred and collected were identified as B. tabaci, twenty-five data sets were processed, whereas 100% of the whiteflies based on visits and interviews with 274 collected in the tropical highlands were farmers and 168 extension agents, T. vaporariorum. The species co-exist in private technical assistants, insecticide all mid-altitude valleys where samples salesmen and government officers in were taken, with T. vaporariorum charge of rural development projects. predominating. Thus, in Colombia and Insecticide resistance was also Ecuador, T. vaporariorum occurs assessed in whiteflies from 40 sites, as between 780 and 2830 m above sea detailed in Chapter 4.3 (this volume). level and is the only species present above 1600 m, whereas B. tabaci is the Whiteflies and natural enemies only species present at lower elevations were identified at CIAT and whitefly (500 m or less) (Quintero et al., 2001). biotypes determined using esterase banding patterns in polyacrylamide gel This distribution of the two major electrophoresis (PAGE) and random whitefly species may be related to amplified polymorphic DNA/ climatic conditions, especially polymerase chain reaction (RAPD-PCR) temperature regimes, which are analysis (Quintero et al., 1998). determined in the tropics principally by Biological samples for whitefly species elevation. As much of the agricultural and biotype identification were taken at production in the Andes is 215 sites; 74 samples for identification concentrated in mid-altitude valleys, of natural enemies were collected. our findings clearly show the need for a correct identification of the whitefly Whitefly populations were sampled species concerned. If it is assumed, for in 14 crops. In the tropical highlands, example, that B. tabaci is the major emphasis was placed on common bean pest in mid-altitude valleys and (73% of samples taken), tomato (18%) highlands, IPM packages will be ill and potato (7%). In mid-altitude targeted. Our findings also dispel the valleys, most samples were taken on notion, prevalent among technical tomato (58%) and common bean (18%). Cucurbits (38%), tomato (31%), pepper (Capsicum annuum L.) (10%) and 100 100 91.5 eggplant (Solanum melongena L.) (5%) were the crops most intensively 80 60.8 surveyed in the tropical lowlands. Yield 60 losses due to the greenhouse whitefly 40 on snap beans were measured. 25.5 20 13.7 8.5 Whitefly species (%) 0 0 0 0 Increased Biological Tropical Mid-altitude Tropical lowlands valleys (400- highlands Understanding (<400 masl) 1000 masl) (>1000 masl) Bemisia Bemisia Trialeurodes Species tabaci A tabaci B vaporariorum An analysis of whitefly species Figure 2. Composition of whitefly populations composition along an altitudinal in three major ecological zones, as defined by altitude (masl = metres gradient suggests that T. vaporariorum above sea level), in Colombia and and B. tabaci occupy well-defined Ecuador.

276 Colombia and Ecuador assistants and extension personnel, B. tabaci was detected only in four that B. tabaci is the key pest in the departments in Colombia: Córdoba, highlands and mid-altitude valleys of Sucre, Valle and Huila (Figure 2). The the Andean zone. T. vaporariorum is the prevalence of the B biotype in the key pest in this region. coastal areas of Ecuador and Colombia represents a serious threat Biotypes to agriculture in the region. Recent work (CIAT, unpublished results) has No evidence for the existence of shown that the presence of the B different biotypes of T. vaporariorum biotype can also be a major threat to was found in more than 400 samples agriculture in mid-altitude valleys examined by RAPD-PCR. Thus it is such as the Cauca Valley in concluded that there is no basis for Colombia. In a survey conducted in continued biotype analysis of the 2003, the B biotype was present in greenhouse whitefly within the context 63% of samples taken (as opposed to of the TWF-IPM Project. 11.5% in 1997). The increased incidence of the B biotype was also Both A and B biotypes of B. tabaci evidenced by the common were detected among the 214 samples occurrence of such physiological evaluated by RAPD-PCR (Figure 3). The disorders as irregular ripening of B biotype is the only one of importance tomato, and silver leaf symptoms on along the coast of Ecuador. It is squash (Cucurbita moschata present also throughout the northern Duchesne). This underlines the need (Atlantic) coast of Colombia and in the for continuous monitoring of whitefly Departments of Valle, Huila, Tolima species composition. and Cundinamarca. The A biotype of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 M

Tv A B B A Tv 100 B. tabaci biotype B

Figure 3.hk RAPD patterns obtained with primer OPA-04 for a sample of whiteflies Chcollected k in the northern coastal zone of Colombia. Lane 1: Trialeurodes vaporariorum (control); lane 2: Bemisia tabaci biotype A (control); lane 3: B. tabaci biotype B (control); lanes 4-12: B. tabaci biotype B (field sample); lane 13: B. tabaci biotype B (control); lane 14: B. tabaci biotype A (control); lane 15: T. vaporariorum (control); lane M: marker (1 kb ladder).

277 Whiteflies and Whitefly-borne Viruses in the Tropics

Reproductive hosts Matsum. & Nakai), pumpkin (Cucurbita ficifolia Phaseolus In Colombia, T. vaporariorum was Bouché), lima bean ( lunatus L. var. lunatus), forage Arachis found breeding on common bean and Arachis pintoi snap bean, tomato, potato, squash, ( Krapov. & W. Gregory), Vitis vinifera cucumber (Cucumis sativus L. var. grape ( L.), poinsettia and additional Brassica species. Table 1 sativus), eggplant, pepper, pea (Pisum sativum L.), tobacco (Nicotiana tabacum shows the main crops affected by B tabaci L.), cabbage (Brassica oleracea L.), . . coriander (Coriandrum sativum L.) and several species of ornamentals. In mid- In greenhouse studies on the altitude areas, T. vaporariorum also comparative biology of biotypes A and reproduces on cotton and soybean. All B on 20 cultivated host plants, the T. vaporariorum samples in Ecuador B biotype bred faster than the were taken on common bean and A biotype on cauliflower (Brassica tomato. Snap bean and common bean oleracea L. var. botrytis L.), cabbage, are by far the most important squash and tomato. Percentage reproductive hosts for T. vaporariorum emergence of the B biotype was in Colombia and Ecuador, followed in significantly higher on 13 out of the order of decreasing importance by 20 host plants studied; the A biotype tomato and potato (Table 1). did not breed on squash or beet (Beta vulgaris L. subsp. vulgaris). The B In Colombia, the A biotype of biotype did more damage to lettuce, B. tabaci was recorded breeding on squash, cabbage, cauliflower, soybean, tobacco, tomato, cotton, cucumber, Brassica sp., soybean and poinsettia (Euphorbia pulcherrima Willd. watermelon than did the A biotype. ex Klotzch), broccoli (Brassica oleracea Snap bean was affected also and L. var. italica Plenck) and eggplant. In suffered significant reductions in both Colombia and Ecuador, the biomass as a result of damage by the B B biotype was found reproducing on biotype. Common bean, on the other cotton, tomato, eggplant, squash, hand, was not as severely affected as pepper, cucumber, common bean, other host plants. snap bean, soybean, tobacco, cabbage, lettuce (Lactuca sativa L. var. capitata The B biotype successfully L.), sweetpotato (Ipomoea batatas [L.] colonized plants of four Cucurbita Lam.), melon (Cucumis melo L.), species tested. The A biotype did not watermelon (Citrullus lanatus [Thunb.] survive to the adult stage on pumpkin

Table 1. Main crops serving as reproductive hosts of whiteflies in major agro-ecological zones of Colombia and Ecuador.

Whitefly species Tropical lowlands Mid-altitude valleys Tropical highlands

A biotype of Bemisia tabaci Eggplant Tobacco - Tomato Soybean - B biotype of B. tabaci Squash Squash - Cotton Cotton - Eggplant Tobacco - Brassica sp. - - Tomato - - Trialeurodes vaporariorum - Common bean Common bean - Snap bean Snap bean - Tomato Tomato - - Potato

278 Colombia and Ecuador and squash. On Cucurbita mixta lecanii (Zimm.) and Paecilomyces Pangalo (= Cucurbita argyrosperma C. fumosoroseus (Wize) were found in the Huber subsp. argyrosperma) and Departments of Antioquia, Cauca Cucurbita pepo L., the B biotype Valley and Cundinamarca in Colombia reached the adult stage in ca. 20 days, (López-Avila et al., 2001). 33% faster than the A biotype. The ability of the B biotype to successfully Additional studies by Manzano et colonize hosts that are not attacked by al. (2000) demonstrated that the A biotype and its ability to A. fuscipennis is abundant on common reproduce faster on certain host plants bean crops where no chemical are important diagnostic tools (Brown treatments have been made, and et al., 1995; Schuster et al., 1996). remains active even after heavy When C. moschata, C. ficifolia and pesticide use. Some studies have been C. pepo plants were infested with either conducted on the biology or efficacy of A or B biotype adults, full expression of this species as a natural enemy of silverleaf symptoms, as described by T. vaporariorum in the Andean zone Yokomi et al. (1990) and Costa and (Márquez and Valencia, 1991; Medina Brown (1991), were obtained 21 days et al., 1994; Manzano et al., 1999). after infestation with the B biotype. No Recent results (Manzano et al., 2000) silvering was observed on any plant suggest that A. fuscipennis could infested with the A biotype. potentially be a good biological control agent of T. vaporariorum in Colombia, Apart from silverleaf, other in environments that are not overly dry symptoms of B biotype attack that or warm. However, as the parasitoid have been described in the literature seems to be negatively affected by were observed in numerous different certain combinations of climatic fields: uneven ripening of tomato conditions, its limitations need to be (Schuster et al., 1996; Polston and further studied in order to ascertain its Anderson, 1997), chlorotic pods and true potential as a natural enemy. petioles on snap beans (Hassan and Sayed, 1999), white stem streaking of The entomopathogen V. lecanii has Brassica spp. and phytotoxic disorders been tested under laboratory and field on poinsettia and lettuce (Brown et al., conditions in Colombia (González and 1995). López-Avila, 1997) with variable, usually poor, results. Natural enemies The natural enemies of B. tabaci, Survey work confirmed that Amitus found at many different sites in fuscipennis MacGown and Nebeker is Colombia and Ecuador, included the the most common parasitoid of aphelinid parasitoids E. hispida, T. vaporariorum in the highlands of E. pergandiella, E. nigricephala, Colombia and Ecuador. Other natural Encarsia sp. and Eretmocerus sp., as enemies of T. vaporariorum recorded well as the predators D. pusillus and were the parasitoids Encarsia hispida Chrysoperla sp. The relative importance DeSantis, Encarsia pergandiella and efficiency of these natural enemies Howard, Encarsia nigricephala Dozier has not been assessed. and Encarsia sp., and the predators Delphastus pusillus Leconte, Hippodamia convergens (Guérin- Yield losses Ménenville), Cycloneda sanguinea (L.), Three trials aimed at measuring losses Coleomegilla maculata (DeGeer) and due to T. vaporariorum on snap bean Chrysoperla sp. The fungi Verticillium were conducted during 1998 and 1999.

279 Whiteflies and Whitefly-borne Viruses in the Tropics

The establishment of these trials distributed extensively, the insects coincided with the appearance in being present in 72% of farms visited. Colombia of another serious pest of The lowest incidence was in the common bean, the melon thrips, Thrips Departments of Cauca and Cauca palmi Karny. As a result, the trials had Valley, Colombia, where the insects to be redesigned to account for the were present in 46% of farms visited. incidence of the new pest. In large replicated plots, different insecticide Whiteflies ranked as the most regimes were established to measure important pest of common bean in yield responses and to partition losses 50% of the sites visited, and as the key due to thrips and whiteflies. The latter pest in 45% of the tomato farms that was achieved by using spinosad were surveyed. A relatively small (Tracer®) Dow Chemical Co.) as a percentage of farmers gave whiteflies selective insecticide for T. palmi control the highest rating as pests when and buprofezin (Applaud®, ICI) in compared with other insects on the mixture with monocrotophos farm (for example, T. palmi on common (Azodrin®, Shell Co.) for whitefly bean or fruit borers on tomato) but a exclusion. The mean loss due to T. significant proportion thought that palmi was 4.1 t/ha or 46% of the whiteflies could reduce yields by 50% potential yield. T. vaporariorum caused or more. Up to 30% of melon growers losses of 2.1 t/ha or 23% of potential and 34% of those planting tomato yield (Rendón et al., 2001). indicated that they have abandoned crops as a result of devastating whitefly attacks. Most farmers (76% of those Increased Socio-Economic growing cucurbits, 80% of those Understanding growing tomato) could identify the insects as pests but very few associated the presence of whiteflies with virus Farmers’ perceptions of the diseases (Table 2). whitefly problem From 88% to 100% of farmers Technicians’ perceptions of the surveyed were male; 41% to 63% were whitefly problem owners who had been on their farms Virtually all extension agents and for over 5 years, planting an average of technical assistants interviewed 1.2 ha with three or more crops considered that whiteflies were a major planted in succession through the year. problem on tomato and common bean. Table 2 shows that most farmers knew Most (80%-93%) regarded whiteflies as the insects; they were able to identify pests, while 27%-39% considered that them as whiteflies and seemed to be whiteflies act as both pests and aware of whitefly-related problems. The vectors. Up to 70% thought that the whitefly was regarded as an endemic situation in most crops had changed as pest by about 49% of farmers in the a result of whitefly incidence. Very few tropical highlands of Colombia and were aware of new technologies to solve Ecuador, by 23% in mid-altitude the problem, considered that the area valleys (400-1000 m) in Colombia and planted to several crops had been by 82% in the tropical lowlands of reduced or that insecticide use had Colombia and Ecuador. A minority increased as a result of whitefly claimed to be able to predict whitefly incidence. All acknowledged that using epidemics, which they usually insecticides was the only associated with dry weather. Whitefly recommendation they could make to problems in the survey area are farmers.

280 Colombia and Ecuador

Table 2. Main characteristics and perceptions of farmers surveyed in the whitefly-affected areas of Colombia and Ecuador.

Percentage of farmers who: Tropical Mid-altitude Tropical highlands valleys lowlands

Are male 88 100 100 Own their land 63 41 64 Have lived on the farm for more than 5 years 64 41 68 Receive technical assistance 15 18 42 Plant three or more crops simultaneously 47 68 36 “Rotate”a crops and/or fields 82 79 59 Plant more than twice a year 56 64 11 Know whiteflies 82 92 88 Regard whiteflies as a problem 67 64 65 Regard whiteflies as endemic 49 23 82 Relate whitefly incidence to climatic conditions 53 59 83 Consider that they can predict whitefly incidence 24 14 53 Give highest rating to whiteflies as pests 15 18 21 Consider that whiteflies can cause 50% or more yield losses 40 14 40 Identify whiteflies as pests 71 32 80 Identify whiteflies as virus vectors 0 5 0 Identify whiteflies as both pests and vectors 1 9 7 a. Rotation in this case sometimes means shifting the crop from one place on the farm to another.

Insecticide use who recommended a wide array of insecticides (up to 34 different brands) All farmers who attempted control of whiteflies (up to 70%) in mid-altitude for whitefly control. The three crops most heavily sprayed were tomato, and highland areas of Colombia and common bean and potato but, as Ecuador used insecticides as the sole method of control. Virtually all shown in Table 3, many applications did not have whiteflies as the main applications against whiteflies were target insect. Nevertheless, the heavy preventive, with little or no regard for insect populations and the use of insecticides against other pests favours the development of the high phenological stage of crops. Few levels of resistance detected in farmers received formal technical assistance; instead, they received whiteflies in the Andean zone (Chapter 4.3, this volume). frequent visits by insecticide salesmen

Table 3. Percentage of farmers using insecticides to control whiteflies and other pests on selected crops in major ecological zones of Colombia and Ecuador.

Crop Ecological zone Using Spraying against: insecticides Whiteflies Whiteflies and Other other insects insects

Common bean Tropical highlands 89 33 39 17 Common bean Mid-altitude valleys 75 0 75 0 Tomato Tropical highlands 90 11 63 16 Tomato Mid-altitude valleys 92 8 50 34 Tomato Tropical lowlands 86 38 48 0 Potato Tropical highlands 86 0 0 86

281 Whiteflies and Whitefly-borne Viruses in the Tropics

Up to 57% of insecticide toxicological category I (extremely applications were reportedly made with dangerous), as defined by Metcalf organophosphate insecticides (OPs) (1994) and are used with minimal alone. However, OPs also were reported safety precautions for handling and as being used in mixtures with application. In the survey, the most carbamates and pyrethroids (Table 4) widely used group of pesticides were to the extent that 80%-100% of the OPs, with 40% to 50% share of the applications had OPs as ingredients. use, followed by carbamates and Given the high levels of insecticide pyrethroids (Table 5). Novel products resistance to OPs and pyrethroids that such as insect growth regulators, were detected in our studies, it is not juvenoids and nicotinoids were surprising that a significant proportion reported as becoming popular of farmers (up to 60%) were beginning (Rodríguez and Cardona, 2001). Also to use novel insecticides. This was used were soaps, oils and botanicals. most evident in tomato-growing areas in the tropical lowlands and in the mid- Insecticides were reportedly applied altitude valleys of Colombia. to tomato and snap bean as often as three times a week (Rodríguez and Farmers had easy access to at least Cardona, 2001). The number of 36 different active ingredients and applications per cropping season varied 40 commercial brands. Most among regions but it is interesting to insecticides used belong to the note that 24% of farmers surveyed made 10 or more applications (Table 6).

Table 4. Percentage of farmers using Most farmers used very low organophosphates (OP), carbamates (0.2 cc/L) or very high (4.8 cc/L) (CAR) and pyrethroids (PYR) to dosages of concentrate, well below or control whiteflies in major ecological above the 2 cc/L average dosage zones of Colombia and Ecuador. usually recommended by Insecticide Tropical Mid- Tropical manufacturers. Up to 71% of farmers group highlands altitude lowlands valleys surveyed in Colombia and 55% in Ecuador took their own decisions on OP 55 33 57 timing of applications as opposed to CAR 4 0 2 those who have access to technical PYR 4 0 8 assistance. More than half reported OP + CAR 5 45 5 spraying on a calendar basis, OP + PYR 24 22 25 regardless of infestation levels, and CAR + PYR 0 0 0 relatively few had received any kind of OP + CAR + PYR 8 0 3 technical assistance.

Table 5. Percentage participation of organophosphate, carbamate and pyrethroid insecticides in chemical control of whiteflies in selected tropical highland areas of Ecuador and Colombia.

Insecticide category Northern Ecuador– Departments of Department of southern Colombia Cauca and Cauca Valley, Antioquia, Colombia Colombia

Organophosphates 40 50 50 Carbamates 22 14 25 Pyrethroids 22 9 13 Novel productsa 16 23 12 Others (soap, oils, etc.) 0 4 0 a. Includes insect growth regulators, juvenoids and nicotinoids.

282 Colombia and Ecuador

Table 6. Patterns of insecticide use for Costa, H. S.; Brown, J. K. 1991. Variation whitefly control in Colombia and in biological characteristics and Ecuador. esterase patterns among Number of insecticide Percentage populations of Bemisia tabaci, and applications per cropping of farmers the association of one population seasona with silverleaf symptom induction. 1-3 34.8 Entomol. Exp. Appl. 61:211-219. 4-6 25.0 González, J. G.; López-Avila, A. 1997. 7-9 8.8 Evaluación de cepas nativas de 10 7.4 Verticillium lecanii (Zimm.) en el > 10 24.0 control de la mosca blanca de los invernaderos Trialeurodes a. The mean number of applications per season against the B biotype of Bemisia vaporariorum (Westwood). Rev. tabaci in the tropical lowlands of Colombia Colomb. Entomol. 23(1-2):25-30. and Ecuador is 6.5. Hassan, A. A.; Sayed S. F. 1999. Chlorotic pod: A new physiological disorder of Conclusions green-podded snap beans (Phaseolus vulgaris L.) associated In general, the results of both the with silverleaf whitefly infestation. Egypt. J. Hortic. 26:213-228. biological surveys and the interviews with producers and their technical López-Avila, A.; Cardona, C.; García, J.; advisors underline the need for IPM Rendón, F.; Hernández, P. 2001. research and implementation. A highly Reconocimiento e identificación de dangerous situation has developed for enemigos naturales de moscas small-scale farmers in Colombia and blancas (Homoptera: Aleyrodidae) en Colombia y Ecuador. Rev. Ecuador over the past 15 years: Colomb. Entomol. 27(3-4):137-141. commercial pressures and an inadequate knowledge base have Manzano, M. R.; van Lenteren, J. C.; influenced farmers to adopt a Cardona, C. 1999. Some biological “chemical culture”, involving excessive characterisitics of Amitus dependence on pesticides. The fuscipennis MacGown and Nebeker situation has been aggravated by the (Hymenoptera: Platygasteridae), parasitoid of the greenhouse introduction into the region of the B whitefly. IOBC Bull. 23(1):169-172. biotype of B. tabaci, a major pest of annual crops. The results, their Manzano, M. R.; van Lenteren, J. C.; consequences and recommendations Cardona, C.; Drost, Y. C. 2000. for further action are discussed in Developmental time, sex ratio, and more detail in Chapter 4.4. longevity of Amitus fuscipennis MacGown and Nebeker (Hymenoptera: Platygasteridae) on the greenhouse whitefly. Biol. References Control 18(2):94-100.

Brown, J. K.; Frohlich, D. R.; Rosell, R. C. Márquez, M. L.; Valencia, S. A. 1991. 1995. The sweetpotato or silverleaf Evaluación de Encarsia formosa whiteflies: Biotypes of Bemisia Gahan y Amitus fuscipennis tabaci or a species complex? Ann. MacGown and Nebeker, en el Rev. Entomol. 40:511-534. control de Trialeurodes vaporariorum (Westwood) en crisantemo (Chrysanthemum morifolium Rainat.). B.Sc. thesis, Universidad Nacional de Colombia, Medellín, CO. 132 p.

283 Whiteflies and Whitefly-borne Viruses in the Tropics

Medina, P. S.; Saladarriaga, A. V.; Pérez, Rendón, F.; Cardona, C.; Bueno, J. 2001. G. 1994. Biología de Amitus Pérdidas causadas por Trialeurodes fuscipennis MacGown and Nebeker, vaporariorum (Homoptera: bajo tres condiciones ecológicas, en Aleyrodidae) y Thrips palmi Rionegro (Antioquia). Rev. Colomb. (Thysanoptera: Thripidae) en Entomol. 20(3):143-148. habichuela en el Valle del Cauca. Rev. Colomb. Entomol. 27(1-2): Metcalf, R. L. 1994. Insecticides in pest 39-43. management. In: Metcalf, R. L.; Luckmann, W. H. (eds.). Rodríguez, I.; Cardona, C. 2001. Introduction to insect pest Problemática de Trialeurodes management. 3rd edition. John vaporariorum y Bemisia tabaci Wiley, NY, USA. p. 245-284. (Homoptera: Aleyrodidae) como plagas de cultivos semestrales en el Polston, J. E.; Anderson P. K. 1997. The Valle del Cauca. Rev. Colomb. emergence of whitefly-transmitted Entomol. 27(1-2):21-26. geminiviruses in tomato in the western hemisphere. Plant Dis. Schuster, D. J.; Stansly, P. A.; Polston J. 81:1358-1369. E. 1996. Expressions of plant damage by Bemisia. In: Gerling, D.; Quintero, C.; Cardona, C.; Ramírez, D.; Mayer, R.T. (eds.). Bemisia: 1995. Jiménez, N. 1998. Primer registro Taxonomy, biology, damage, control del biotipo B de Bemisia tabaci and management. Intercept, (Homoptera: Aleyrodidae) en Andover, GB. p. 153-165. Colombia. Rev. Colomb. Entomol. 24(1):23-28. Yokomi, R. K.; Hoelmer, K. A.; Osborne, L. S. 1990. Relationships between the Quintero, C.; Rendón, F.; García, J.; sweetpotato whitefly and the Cardona, C.; López-Avila, A.; squash silverleaf disorder. Hernández, P. 2001. Especies y Phytopathology 80:895-900. biotipos de moscas blancas (Homoptera: Aleyrodidae) en cultivos semestrales de Colombia y Ecuador. Rev. Colomb. Entomol. 27(1-2):27-31.

284 Insecticide Resistance in Colombia and Ecuador

CHAPTER 4.3 Insecticide Resistance in Colombia and Ecuador

César Cardona*, Francisco Rendón*, Isaura Rodríguez* and Aristóbulo López-Avila**

Farmer interviews conducted as part of results. However, the 10 most widely extensive on-farm surveys for the used insecticides identified in the diagnostic phase of the Tropical surveys comprised nine conventional Whitefly Integrated Pest Management products—dimethoate, carbofuran, (TWF-IPM) Project (Rodríguez and chlorpyriphos, methamidophos, Cardona, 2001; Chapter 4.1, this methomyl, profenofos, monocrotophos, volume) showed that insecticide use cypermethrin and malathion—and only against whiteflies in Colombia and one of these novel insecticides, Ecuador is excessive. In the tropical imidacloprid. The high toxicity of highlands and mid-altitude valleys, several of the conventional products in farmers spray their crops 5 to 6 times widespread use raises concerns over on average to control Trialeurodes both human and environmental health, vaporariorum (Westwood). The mean underlining the need for alternative number of applications against the B approaches based on IPM. biotype of Bemisia tabaci (Gennadius) in the tropical lowlands of Colombia Insecticide resistance in whiteflies and Ecuador was estimated at is well documented as a widespread 6.5. Over-reliance on insecticides for phenomenon in those countries in whitefly control is so widespread that which monitoring of resistance has 30% of 325 farmers interviewed been conducted (Horowitz and Ishaaya, reported that they make more than 1996). Recent reviews on the subject 10 applications per cropping season. are those of Dittrich et al. (1990), Cahill The frequency of applications in many et al. (1996b), Denholm et al. (1996) cases is as high as two to three times and Palumbo et al. (2001). In the per week. Most farmers complained Andean zone, Buitrago et al. (1994) about the limited control achieved by detected high levels of resistance to conventional insecticides and many cypermethrin and deltamethrin in farmers are now using novel populations of T. vaporariorum from the insecticides such as buprofezin, central highlands of Colombia. pyriproxyfen, diafenthiuron and Resistance to organophosphates and imidacloprid reportedly with better carbamates was not as high, and ranged from low to moderate. Cardona et al. (1998; 2001) reported that the B * Centro Internacional de Agricultura Tropical biotype of B. tabaci, the main whitefly (CIAT), Cali, Colombia. species in the lowlands of Colombia, ** Corporación Colombiana de Investigación Agropecuaria (CORPOICA), Bogotá, was highly resistant to methamidophos Colombia. and methomyl.

285 Whiteflies and Whitefly-borne Viruses in the Tropics

A more general account of whitefly Laboratory work to establish problems in the high Andes, mid- baseline data and select diagnostic altitude valleys and lowlands of dosages was conducted at the Colombia and Ecuador is presented headquarters of the Centro elsewhere (Chapter 4.2, this volume), Internacional de Agricultura Tropical as are the results of biological surveys (CIAT) in Cali, Colombia. Average and farmer questionnaires, intended to laboratory conditions were 24 °C and characterize whitefly problems in these 75%-80% relative humidity (RH). areas. Here, a summary is presented of findings on the resistance of whiteflies Field monitoring extended from to conventional insecticides in the northern Ecuador (Imbabura and survey areas. Also included are Carchi provinces) to the north-eastern baseline responses of T. vaporariorum department of Guajira in Colombia. adults to monocrotophos, lamda- With the exception of coastal areas in cyhalothrin, bifenthrin, carbosulfan, Ecuador, all major areas affected by carbofuran, thiamethoxam and whiteflies were covered. Sites were imidacloprid. Baseline responses of T. chosen according to previous data on vaporariorum nymphs to buprofezin, heavy insecticide use but care was diafenthiuron and imidacloprid are taken not to perform tests in areas with included. The knowledge of whitefly mixed populations of whitefly species, problems synthesized in these studies as indicated in biological surveys is intended to provide a sound basis for carried out in parallel (Chapter 4.2, pursuing IPM approaches in the this volume). Thus, no insecticide Andean region of Latin America. resistance work was conducted in mid- altitude valleys where mixed populations of B. tabaci and Methods and Materials T. vaporariorum had been recorded. In total, insecticide resistance was measured at 40 sites in Colombia and Experimental approach and Ecuador. Between two and four locations insecticide resistance tests were Field surveys and insecticide resistance conducted in each target area. testing were carried out between Resistance of T. vaporariorum collected October 1997 and May 1999. In from common bean (Phaseolus vulgaris accordance with the common research L.), tomato (Lycopersicon esculentum L.) methods agreed among the project and potato (Solanum tuberosum L.) was partners, the vial technique was measured in highland areas along the adopted to establish baseline data in Andean corridor from northern the laboratory and for field assessment Ecuador to northern Colombia. of resistance to conventional Resistance of B. tabaci biotype B from insecticides. The technique involves tomato, eggplant (Solanum melongena assessing the rate of mortality of a L.), squash (Cucurbita moschata sample of whiteflies within a vial coated Duchesne) and cotton (Gossypium with a known concentration of hirsutum L.) was measured in lowland insecticide (Plapp et al., 1990; Cahill areas of the region known as the and Hackett, 1992). Baseline data for “northern coast” of Colombia and in imidacloprid were obtained following the Cauca Valley (a mid-altitude valley). the methodology developed by Cahill et al. (1996a).

286 Insecticide Resistance in Colombia and Ecuador

Baseline data and diagnostic recorded 6 hours after introduction of doses the adults into the units. Adult whiteflies obtained from separate mass rearings of T. vaporariorum and To generate baseline data with B. tabaci biotype A, maintained at CIAT nymphs, we collected common bean leaves infested with pupae in each of for over 10 years (without exposure to insecticides) were used to obtain the test sites. The infested material was baseline data. Technical grade then confined to cages. Adults emerging from these pupae were insecticides were dissolved in acetone and 250 µL samples of the solution collected with an aspirator and were pipetted into each 20-mL vial. The introduced in clip cages attached to leaves of common bean seedlings. The acetone was allowed to evaporate completely before introducing the test adults (10 per clip cage) were allowed insects. Insecticides and dosages tested to oviposit for 24 hours. After hatching of the eggs, the area of the leaf infested with T. vaporariorum were methamidophos (16.0, 8.0, 4.0, 2.0, with first instar nymphs was marked 1.0 and 0.5 µg/vial), methomyl and the number of individuals thus obtained was recorded. Infested (2.5, 0.5, 0.1 and 0.02 µg/vial), cypermethrin (100.0, 50.0, 25.0, seedling leaves were then dipped for 12.5 and 6.25 µg/vial) and carbofuran 5 s in 100 mL of the desired concentration of each of the (10.0, 3.3, 1.1, 0.33 and 0.11 µg/vial). Insecticides and dosages tested with insecticides tested. Formulated the A biotype of B. tabaci were material was used in all tests and the desired concentrations were prepared methomyl (25.0, 6.25, 1.56, 0.39 and 0.10 µg/vial), methamidophos (16.0, by dilution with distilled water. 4.0, 2.0, 1.0 and 0.5 µg/vial) and Controls were dipped in distilled water. Treated plants were kept in a rearing cypermethrin (450.0, 150.0, 50.0, o 16.7 and 5.6 µg/vial). Five replicates of chamber at 24 C, 75%-80% RH and 20 insects were tested for each dosage. left undisturbed until adult emergence occurred. The number of surviving Acetone-coated vials were used as checks (controls). Mortality was individuals in each treatment was recorded 6 hours after the adults were recorded. Insecticides and dosages tested were buprofezin (100, 25, 6.2, placed in the vials. 1.5, 0.3, 0.09 and 0.02 ppm), To generate baseline data for diafenthiuron (1000, 300, 100, 30, 10, 3, 1 and 0.3 ppm) and imidacloprid, imidacloprid and thiamethoxam (Rodríguez et al., 2003), petioles of (1000, 300, 100, 30, 10, 3 and 1 ppm). excised common bean leaves were Four replicates were tested for each dosage. immersed in aqueous solutions of commercial grade imidacloprid (10.0, 5.0, 2.5, 1.25, 0.625 and 0.312 ppm) Percentage mortality was corrected using the Abbott formula (Busvine, for 48 hours to allow the leaves to take up the insecticide solution. Leaf discs 1971). Tests in which check mortality were then cut and placed on agar in was higher than 10% were not included in the analysis. Results were petri dishes, one leaf disk per petri dish (Cahill et al., 1996a) and infested with subjected to probit analysis (SAS Institute, 1988). Slopes, LC and LC 20 T. vaporariorum adults per unit. 50 90 values, as well as 95% confidence Discs from leaves whose petioles had been immersed in distilled water were limits were tabulated. Baseline data used as checks. Five replicates (units) were then used to choose diagnostic doses. Four doses were chosen per dosage were used. Mortality was

287 Whiteflies and Whitefly-borne Viruses in the Tropics empirically so as to obtain mortality ice-chests for 6 hours before mortality ranging from 5% to 95%. After several was recorded. Any test in which tests, a single dose for each insect mortality in control vials was higher species and insecticide was chosen as than 10% was discarded. Percentage the diagnostic dose (as defined by mortality was corrected using the Halliday and Burnham, 1990) to be Abbott formula (Busvine, 1971) and used in extensive monitoring of transformed to arcsine square root of resistance under field conditions. proportion. Data were then analysed by a 1-way analysis of variance (ANOVA) Field assessment of resistance (SAS Institute, 1988). When the F test was significant, comparison of means Resistance was assessed in the field was by the least significance difference using vials coated in the laboratory (LSD). Untransformed means are with diagnostic dosages of methomyl, presented. methamidophos and cypermethrin, and transported to the field in ice-chests maintained at about 20 °C. These insecticides are representative of the Results and Discussion conventional insecticides most widely used in the study area. At each site, Baseline data and diagnostic adult whiteflies were collected with a doses mouth aspirator from the foliage of infested plants and transferred to the Table 1 presents baseline toxicology vials. Twenty adults per vial, replicated data for reference strains of five times, were used for each test. T. vaporariorum and Table 2, the A

Acetone-coated vials were used as biotype of B. tabaci. The LC50 and LC90 controls. Infested vials were kept in the values reflect toxicities of the test

Table 1. Toxicological responses of laboratory strains of Trialeurodes vaporariorum to five insecticides.a

b b 2 Insecticide n LC50 (95% FL) LC90 (95% FL) Slope ± SEM X Adults Methomyl 457 0.25 (0.15-2.6) 0.95 (0.76-14.7) 2.19 ± 0.55 4.05 Methamidophos 600 5.30 (2.5-7.6) 22.50 (15.6-48.7) 2.05 ± 0.49 0.08 Monocrotophos 710 9.70 (6.7-13.4) 175.40 (115.5-299.8) 1.00 ± 0.08 4.80 Cypermethrin 480 37.00 (22.0-55.7) 400.00 (232.7-953.5) 1.24 ± 0.18 2.95 Lambda-cyhalothrin 605 9.40 (6.1-13.5) 264.90 (170.9-455.3) 0.90 ± 0.06 0.10 Bifenthrin 380 2.40 (1.6-3.1) 6.70 (5.2-9.8) 2.90 ± 0.49 0.90 Carbofuran 504 1.97 (1.5-2.5) 6.80 (5.4-9.7) 2.37 ± 0.31 4.31 Carbosulfan 712 1.80 (1.5-2.1) 19.90 (16.3-24.9) 1.20 ± 0.05 3.30 Imidacloprid 921 5.70 (4.6-6.9) 28.40 (20.9-44.9) 1.90 ± 0.22 7.80 Thiamethoxam 670 8.60 (6.0-11.7) 101.00 (68.8-170.4) 1.20 ± 0.12 5.90

First instar nymphs Buprofezin 907 0.80 (0.6-1.1) 9.20 (6.9-13.0) 1.20 ± 0.09 7.30 Diafenthiuron 904 3.20 (2.3-4.4) 60.10 (41.8-92.8) 1.00 ± 0.06 10.50 Imidacloprid 483 16.50 (10.7-23.4) 171.50 (115.6-290.9) 1.21 ± 0.10 3.60 a. Conventional insecticides were tested using insecticide-coated glass vials. Imidacloprid tests were conducted using the technique developed by Cahill et al. (1996a). Nymphs were tested using a modification of the methodology described by Prabhaker et al. (1985). b. Imidacloprid, thiamethoxam, buprofezin and diafenthiuron concentrations in ppm. All others in µg/vial. LC, lethal concentration; FL, fiducial limits.

288 Insecticide Resistance in Colombia and Ecuador

Table 2. Toxicological responses of laboratory strains of Bemisia tabaci biotype A to three insecticides, using insecticide-coated glass vials.

a a 2 Insecticide n LC50 (95% FL) LC90 (95% FL) Slope ± SEM X Methomyl 500 1.7 (1.1-2.3) 9.1 (6.7-13.7) 1.76 ± 0.21 1.73 Methamidophos 517 1.4 (0.9-1.6) 6.6 (5.3-14.7) 1.86 ± 0.47 14.89 Cypermethrin 502 14.4 (5.8-27.2) 202.9 (122.5-352.4) 1.12 ± 0.14 3.27 a. Insecticide concentration: µg/vial. LC, lethal concentration; FL, fiducial limits. insecticides to susceptible strains of of T. vaporariorum (2.4 ppm) was whiteflies that have not been exposed to similar to that obtained by Cahill et al. insecticides for at least 10 years. As (1996a) using susceptible strains of discussed by ffrench-Constant and B. tabaci from Sudan and Pakistan. Roush (1990), establishing baseline data for different insecticides is a The main purpose of the fundamental step in resistance studies preliminary tests was to select because these data will serve as a basis diagnostic doses for methomyl, for future comparisons, which will allow methamidophos and cypermethrin researchers to detect any changes in (Table 3) that were then used to insecticide resistance levels. In addition, compare the insecticide resistance of as explained by Sanderson and Roush test samples of whiteflies from (1992) and Denholm et al. (1996), Colombia and Ecuador with that of the calculation of baseline data permits the fully susceptible laboratory strains. We selection of diagnostic doses, which can did not calculate diagnostic doses by be used more conveniently in extensive statistical means. Instead, diagnostic efforts to detect and monitor resistance doses were chosen empirically to such as the one reported here that was produce >95% mortality. carried out over numerous sites in the Andean zone of Ecuador and Colombia. Field monitoring of resistance Carbofuran was included because in biotype B of B. tabaci granular formulations of this insecticide Responses to insecticides in our are still effective against studies were arbitrarily classified as T. vaporariorum in Colombia (Cardona, follows on the basis of mean 1995). Imidacloprid is a highly efficient percentage mortality: insecticide that is becoming popular in the region. Careful monitoring of the · 0%-50% mortality: high resistance; efficiency of this product will be needed · 50%-80% mortality: intermediate for future resistance management and resistance; and IPM work to be undertaken by the · 80% mortality: low resistance. second phase of the TWF-IPM Project. It is interesting to note that the LC50 value According to this scale, the B for imidacloprid for our reference strain biotype of B. tabaci recently introduced

Table 3. Response (corrected percentage mortality) of Trialeurodes vaporariorum and Bemisia tabaci adults to three insecticides.a

Whitefly species Methomyl Methamidophos Cypermethrin (2.5 µg/vial) (32 µg/vial) (500 µg/vial)

T. vaporariorum 97 99 88 B. tabaci biotype A 99 100 98 a. Diagnostic dosages were tested using insecticide-coated glass vials.

289 Whiteflies and Whitefly-borne Viruses in the Tropics into Colombia showed high levels of exactly where the B biotype whitefly resistance to methomyl and that invaded Colombia originated and of methamidophos and, in some places, the pattern of insecticide use in the moderate levels of resistance to area of origin. It is likely that the cypermethrin (Table 4). Considerable frequency of insecticide resistance was inter-population variation was detected. already high in the pest population at Variability in the response of the B the time it was introduced. biotype to insecticides has been detected also in Arizona Field monitoring of resistance (Sivasupramaniam et al., 1997), in in T. vaporariorum Hawaii (Omer et al., 1993) and in California (Prabhaker et al., 1996). In The responses of T. vaporariorum to general, insecticide resistance levels insecticides were arbitrarily classified reflected insecticide use patterns. For on the same basis as those of B. tabaci example, the very high levels of (0%-50% mortality, high resistance; resistance to acetylcholine esterase 50%-80% mortality, intermediate inhibitors (exemplified here by resistance; and > 80% mortality, low methomyl and methamidophos) in resistance). In the northern Ecuador- different sites in the departments of southern Colombia region (Table 5), Atlántico, Córdoba and Sucre may T. vaporariorum populations were reflect the extensive use of susceptible to methomyl in seven out of organophosphates on cotton, tomato eight sites tested. Generalized and vegetables in the region. Lower resistance to methamidophos (the most levels of resistance to the test widely used insecticide) was found. pyrethroid (cypermethrin) may be due to Extreme cases were those in northern the less frequent use of this type of Ecuador (Turquisal, Ibarra) and in insecticide for whitefly control (Table 4 southern Colombia (El Tambo, in Chapter 4.2, this volume). However, Gualmatán, Funes). Intermediate to low further interpretation of the data is resistance to cypermethrin was detected hampered by the lack of knowledge of in west-central Colombia (Table 6).

Table 4. Response (corrected percentage mortality) of Bemisia tabaci biotype B adults to three insecticides at sites on tropical lowland areas of the northern coast of Colombia.a

Site (department)b Methomyl Methamidophos Cypermethrin (2.5 µg/vial) (32 µg/vial) (500 µg/vial)

Repelón (Atlántico) 42.8 c 2.0 d 92.0 ab Cereté 1 (Córdoba) 54.0 b 1.5 d 87.8 abc Cereté 2 (Córdoba) 12.4 de 0.8 d 68.8 c Ciénaga de Oro (Córdoba) 41.8 bcd 3.0 d 94.6 a Cotorra (Córdoba) 6.8 e 4.2 d 78.2 bc Montería (Córdoba) 5.4 e 2.4 d 87.2 abc Manaure (Guajira) 100.0 a 25.4 b 91.5 ab Corozal 1 (Sucre) 9.8 e 14.0 bc 85.4 abc Corozal 2 (Sucre) 30.0 bcde 4.8 cd 65.2 c Sampués (Sucre) 15.4 cde 16.2 bc 88.8 abc “CIAT” (reference strain) 100.0 a 100.0 a 100.0 a a. Diagnostic dosages were tested under field conditions using insecticide-coated glass vials. Twenty adults per vial replicated five times were used for each test. Means within a column followed by the same letter are not significantly different at the 5% level (LSD test). b. “CIAT” indicates a laboratory population of B. tabaci biotype A maintained for at least 10 years without exposure to insecticides.

290 Insecticide Resistance in Colombia and Ecuador

Table 5. Response (corrected percentage mortality) of Trialeurodes vaporariorum adults to three insecticides in the southern Colombia-northern Ecuador tropical highland region.a

Site (country)b Methomyl Methamidophos Cypermethrin (2.5 µg/vial) (32 µg/vial) (500 µg/vial)

El Tambo (Colombia) 100.0 a 1.8 d 95.2 a Gualmatán (Colombia) 100.0 a 1.8 d 35.2 c Funes (Colombia) 84.2 a 9.0 d 78.0 b Ibarra (Ecuador) 100.0 a 4.0 d 100.0 a Pimampiro (Ecuador) 87.4 a 46.4 bc 100.0 a San Vicente ( Ecuador) 55.6 b 59.0 b 100.0 a Turquisal (Ecuador) 100.0 a 14.0 cd 100.0 a “CIAT” (reference strain) 96.6 a 99.0 a 87.7 b a. Diagnostic dosages were tested under field conditions using insecticide-coated glass vials. Twenty adults per vial replicated five times were used for each test. Means within a column followed by the same letter are not significantly different at the 5% level (LSD test). b. “CIAT” indicates a laboratory population of whiteflies of the same species as the test populations maintained for at least 10 years without exposure to insecticides and used here as a susceptible check for toxicological studies.

Table 6. Response (corrected percentage mortality) of Trialeurodes vaporariorum adults to three insecticides at sites in the Antioquia-Cauca-Valle tropical highland region of west-central Colombia.a

Site (department)b Methomyl Methamidophos Cypermethrin (2.5 µg/vial) (32 µg/vial) (500 µg/vial)

Carmen de Viboral (Antioquia) 96.0 a 55.0 c 69.8 d El Peñol (Antioquia) 90.0 ab 76.4 b 80.2 cd Rionegro (Antioquia) 100.0 a 43.4 c 84.2 bcd La Unión (Cauca) 100.0 a 99.0 a 98.2 a Pescador (Cauca) 62.0 c 2.6 e 67.0 d Cerrito (Valle) 100.0 a 8.2 e 89.0 abc La Cumbre (Valle) 100.0 a 100.0 a 98.0 a Pradera (Valle) 79.4 bc 2.8 e 64.8 d Tenerife (Valle) 99.6 a 3.0 e 72.4 cd “CIAT” (reference strain) 96.6 a 99.0 a 87.8 bcd a. Diagnostic dosages were tested under field conditions using insecticide-coated glass vials. Twenty adults per vial replicated five times were used for each test. Means within a column followed by the same letter are not significantly different at the 5% level (LSD test). b. “CIAT” indicates a laboratory population of whiteflies of the same species as the test populations maintained for at least 10 years without exposure to insecticides and used here as a susceptible check for toxicological studies.

Whiteflies were susceptible to cypermethrin was detected. In the methomyl in all but two of the sites eastern mountain ranges of Colombia studied. Extreme resistance to (Table 7), T. vaporariorum was methamidophos occurred in the Cauca susceptible to methomyl and less Valley (Cerrito, Pradera, Tenerife) and resistant to methamidophos than in in Cauca (Pescador). These are places other parts of the country, possibly where insecticide use on common bean because of lower insecticide use in that and tomato is heaviest. Again, low to area. intermediate resistance to

291 Whiteflies and Whitefly-borne Viruses in the Tropics

Table 7. Response (corrected percentage mortality) of Trialeurodes vaporariorum adults to three insecticides at sites in the central-eastern tropical highland region of Colombia.a

Site (department)b Methomyl Methamidophosc Cypermethrin (2.5 µg/vial) (32 µg/vial) (500 µg/vial)

Boyacá (Boyacá) 100.0 a 95.3 a 89.0 a Tinjacá (Boyacá) 100.0 a 85.6 ab 75.2 abc Bojacá (Cundinamarca) 100.0 a 40.0 c 57.9 cd Pasca (Cundinamarca) 100.0 a - 81.0 ab Fosca (Cundinamarca) 100.0 a 71.6 bc 41.4 d Garzón (Huila) 100.0 a 52.2 bc 88.8 a Rivera (Huila) 100.0 a 40.0 c 88.7 a Abrego (Norte de Santander) 98.4 a 60.3 bc 71.3 abc Lebrija (Santander) 91.5 b 3.0 d 64.4 bc “CIAT” (reference strain) 96.6 a 98.8 a 87.8 a a. Diagnostic dosages were tested under field conditions using insecticide-coated glass vials. Twenty adults per vial replicated five times were used for each test. Means within a column followed by the same letter are not significantly different at the 5% level (LSD test). b. “CIAT” indicates a laboratory population of whiteflies of the same species as the test populations maintained for at least 10 years without exposure to insecticides and used here as a susceptible check for toxicological studies. c. Test was not conducted at Pasca location.

Conclusions Ecuador (Rodríguez et al., 2003). Because whiteflies can also develop In summary, the B biotype of B. tabaci, resistance to novel insecticides recently introduced into Colombia, (Denholm et al., 1996; Horowitz and showed high levels of resistance to Ishaaya, 1996; Elbert and Nauen, methomyl and methamidophos and, in 2000; Palumbo et al., 2001) monitoring some places, moderate levels of of resistance and development of resistance to cypermethrin. In highland baseline data should continue in order and mid-altitude areas of Colombia to facilitate management of whitefly and Ecuador, T. vaporariorum showed populations. The results are further high resistance to methamidophos, discussed and proposals for further intermediate resistance to action based on these results are cypermethrin and low resistance to presented in Chapter 4.4 (this volume). methomyl. In general, levels of insecticide resistance in whiteflies seem to be related to the intensity of References use of the particular kind of insecticide. Baseline responses of Buitrago, N. A.; Cardona, C.; Acosta, A. T. vaporariorum to carbofuran and 1994. Niveles de resistencia a imidacloprid are presented. insecticidas en Trialeurodes vaporariorum (Westwood) (Homoptera: Aleyrodidae), plaga del These results provide a basis for frijol común. Rev. Colomb. Entomol. future monitoring of resistance and for 20(2):109-114. the development of insecticide resistance management strategies. Busvine, J. R. 1971. A critical review of Novel insecticides such as the techniques for testing diafenthiuron, buprofezin, pyriproxyfen insecticides. The Commonwealth Institute of Entomology, and imidacloprid are still effective for Commonwealth Agricultural whitefly control in Colombia and Bureaux, London, GB. 345 p.

292 Insecticide Resistance in Colombia and Ecuador

Cahill, M.; Hackett, B. 1992. Insecticidal Dittrich, V.; Uk, S.; Ernst, H. 1990. activity and expression of pyrethroid Chemical control and insecticide resistance in adult Bemisia tabaci resistance of whiteflies. In: Gerling, using a glass vial bioassay. In: D. (ed.). Whiteflies: Their bionomics, Procs. 1992 Brighton Crop Prot. pest status and management. Conf. Pests and Diseases. Intercept Ltd., Andover, Hants, GB. p. 251-256. p. 263-285.

Cahill, M.; Gorman, K.; Day, S.; Denholm Elbert, A.; Nauen, R. 2000. Resistance of I. 1996a. Baseline determination Bemisia tabaci (Homoptera: and detection of resistance to Aleyrodidae) to insecticides in imidacloprid in Bemisia tabaci southern Spain with special (Homoptera: Aleyrodidae). Bull. reference to neonicotinoids. Pest Entomol. Res. 86:343-349. Manag. Sci. 56(1):60-64.

Cahill, M.; Denholm, I.; Byrne, F. J.; ffrench-Constant, R. H.; Roush, R.T . Devonshire, A. L. 1996b. Insecticide 1990. Resistance detection and resistance in Bemisia tabaci: Current documentation: The relative roles of status and implications for pesticidal and biochemical assays. management. In: Procs. 1996 In: Roush, R. T.; Tabashnik, B. E. Brighton Crop Prot. Conf. p. 75-80. (eds.). Pesticide resistance in arthropods. Chapman and Hall, NY, Cardona, C. 1995. Manejo de Trialeurodes USA. p. 4-38. vaporariorum (Westwood) en frijol en la zona Andina: Aspectos técnicos, Halliday, W.; Burnham, K. P. 1990. actitudes del agricultor y Choosing the optimal diagnostic transferencia de tecnología. Memoria dose for monitoring insect IV Taller Latinoamericano sobre resistance. J. Econ. Entomol. moscas blancas y geminivirus, 83(4):1151-1159. 16-18 octubre 1995, Zamorano, HN. CEIBA 36(1):53-64. Horowitz, A. R.; Ishaaya, I. 1996. Chemical control of Bemisia– Cardona, C.; Rendón, F.; Rodríguez, I. Management and application. In: 1998. Chemical control and Gerling, D.; Mayer, R. T. (eds.). insecticide resistance of whiteflies in Bemisia: 1995. Taxonomy, biology, the Andean zone: A progress report. damage, control and management. In: Procs. International Workshop on Intercept Ltd. Andover, Hants, GB. Bemisia and geminivirus, 7-12 June p. 537-556. 1998, San Juan, PR. p. 1-70. Omer, A. D.; Johnson, M. W.; Tabashnik, Cardona, C.; Rendón, F.; García, J.; B. E.; Costa, H. S.; Ullman, D. E. López-Avila, A.; Bueno, J.; Ramírez, 1993. Sweetpotato whitefly J. 2001. Resistencia a insecticidas resistance to insecticides in Hawaii: en Bemisia tabaci y Trialeurodes Intra-island variation is related to vaporariorum (Homoptera: insecticide use. Entomol. Exp. Appl. Aleyrodidae) en Colombia y 67:173-182. Ecuador. Rev. Colomb. Entomol. 27(1-2):33-38. Palumbo, J.; Horowitz, A.; Prabhaker, N. 2001. Insecticidal control and Denholm, I.; Cahill, M.; Byrne, F. J.; resistance management for Bemisia Devonshire, A. L. 1996. Progress tabaci. Crop Prot. 20(9):739-765. with documenting and combating insecticide resistance in Bemisia. In: Gerling, D.; Mayer, R. T. (eds.). Bemisia: 1995. Taxonomy, biology, damage, control and management. Intercept Ltd. Andover, Hants, GB. p. 577-603.

293 Whiteflies and Whitefly-borne Viruses in the Tropics

Plapp, F. W.; Jackman, J. A.; Rodríguez, I.; Morales, H.; Cardona, C. Campanhola, C.; Frisbie, R. E.; 2003. Líneas base, dosis Graves, J. B.; Lutrell, R. G.; Kitten, diagnóstico y medición periódica de W. F.; Wall, M. 1990. Monitoring resistencia a insecticidas en and management of pyrethroid poblaciones de adultos e inmaduros resistance in the tobacco budworm de Trialeurodes vaporariorum (Lepidoptera: Noctuidae) in Texas, (Homoptera: Aleyrodidae) en el Valle Mississippi, Louisiana, Arkansas del Cauca, Colombia. Rev. Colomb. and Oklahoma. J. Econ. Entomol. Entomol. 29(1):21-27. 83:335-341. Sanderson, J. P.; Roush, R. T. 1992. Prabhaker, N.; Coudriet, D.; Meyerdirk, Monitoring resistance in D. 1985. Insecticide resistance in greenhouse whitefly (Homoptera: the sweetpotato whitefly Bemisia Aleyrodidae) with yellow sticky tabaci (Homoptera: Aleyrodidae). J. cards. J. Econ. Entomol. 85(3): Econ. Entomol. 78:748-752. 634-641.

Prabhaker, N.; Toscano, N. C.; SAS Institute. 1988. 1998 SAS user’s Henneberry, T. J.; Castle, S. J.; guide. SAS Institute, Cary, NC, Weddle, D. 1996. Assessment of two USA. bioassay techniques for monitoring of silverleaf whitefly (Homoptera: Sivasupramaniam, S.; Johnson, S.; Aleyrodidae) in California. J. Econ. Watson, T. F.; Osman, A. A.; Entomol. 89(4):805-815. Jassim, R. 1997. A glass-vial technique for monitoring tolerance Rodríguez, I.; Cardona, C. 2001. of Bemisia argentifolii (Homoptera: Problemática de Trialeurodes Aleyrodidae) to selected insecticides vaporariorum y Bemisia tabaci in Arizona. J. Econ. Entomol. (Homoptera: Aleyrodidae) como 90(1):66-74. plagas de cultivos semestrales en el Valle del Cauca. Rev. Colomb. Entomol. 27(1-2):21-26.

294 Conclusions and Recommendations

CHAPTER 4.4 Conclusions and Recommendations

César Cardona*, Aristóbulo López-Avila** and Oswaldo Valarezo***

When the Systemwide Program on time, an opportunity to survey Integrated Pest Management (SP-IPM) systematically the main crop launched its project for “Sustainable production areas of this ecologically integrated management of whiteflies as complex region and characterize their pests and vectors of plant viruses in whitefly problems, both from a the tropics” (The Tropical Whitefly IPM biological point of view and in terms of Project, TWF-IPM), the nature of farmers’ perceptions of these problems. whitefly problems in the high Andes of Colombia and Ecuador was poorly Scientists and technicians of the understood. Anecdotal reports and participating national research fragmentary scientific studies, mostly organizations, the Instituto Nacional published only in the “grey” literature, Autónomo de Investigaciones suggested that whitefly problems were Agropecuarias (INIAP) in Ecuador and severe and insecticide misuse the Corporación Colombiana de widespread in this socially and Investigación Agropecuaria economically important horticultural (CORPOICA) in Colombia, conducted area (Chapter 4.1, this volume). The field surveys and farmer interviews seriousness of the reported situation between October 1997 and December led partners in the project planning 1998, in close collaboration with their process to establish a sub-project counterparts from the project’s specifically to look at “Whiteflies as coordinating organization, the Centro pests of annual crops in the highlands Internacional de Agricultura Tropical of Latin America”. However, at the (CIAT) (Chapter 4.2, this volume). outset of the project, the species of Researchers at CIAT continued until whiteflies responsible for the problems May 1999 to complete parallel studies was uncertain and the role of of insecticide resistance in whiteflies insecticides and insecticide-resistance from selected sites within the study in aggravating whitefly-related area (Chapter 4.3, this volume). problems was unclear. The first phase of this project provided, for the first The results of these studies (Chapters 4.2 and 4.3, this volume) provide a unique description of whitefly * Centro Internacional de Agricultura problems in Andean agriculture, which Tropical (CIAT), Cali, Colombia. ** Corporación Colombiana de Investigación is not only of inherent scientific value Agropecuaria (CORPOICA), Bogotá, but also will now serve as a sound Colombia. foundation for action to address the *** Instituto Nacional Autónomo de Investigaciones Agropecuarias (INIAP), very considerable agricultural Manabí, Ecuador. production, human health and

295 Whiteflies and Whitefly-borne Viruses in the Tropics environmental problems that the study the main pest species in mid-altitude revealed. Integrated pest management valleys or highland areas, then IPM approaches are likely to play a key role programs will be ill targeted and may in tackling these problems, both within not be effective. the framework of future phases of the current project and beyond. Key Farmer perceptions lessons learned in this study and Whiteflies, in this case T. vaporariorum, proposals for further action are were considered to be the key pest in discussed here and in the final chapter common bean at 50% of the sites on Conclusions (this volume). visited and the key pest in 45% of the tomato farms that were surveyed. A significant proportion of farmers think Conclusions that whiteflies can reduce yields by 50% or more. This perception may well Whitefly identification and be justified. Previous studies (Cardona et al., 1993; Cardona, 1995) and work host crop importance conducted in the first phase of the The greenhouse whitefly Trialeurodes TWF-IPM Project showed that vaporariorum (Westwood) is the most T. vaporariorum may cause losses important whitefly species in highland ranging from 23% to 54% of the and mid-altitude areas of Colombia and potential yield in snap bean (also Ecuador. Although it affects numerous Phaseolus vulgaris L.). There is no horticultural crops and ornamentals, doubt that whiteflies have a high T. vaporariorum is most important as a damage potential and that farmers are pest of common bean (Phaseolus correct in believing that whitefly vulgaris L.) and tomato (Lycopersicon attacks have created numerous esculentum Mill.). This confirms problems and have become a major previous work (Cardona, 1995; constraint to agricultural production in Corredor et al., 1999) on the increasing the region. importance of T. vaporariorum as a major pest in the highlands of the Insecticide use Andes. However, it is now recognized Whiteflies in Colombia and Ecuador that whitefly problems in the area are have become the target of intensive, in more widespread than initially thought. most cases excessive, use of insecticides. This in turn has resulted Altitudinal distribution in the destruction of natural enemies The results of extensive surveys and the creation of new secondary pest showed that T. vaporariorum occurs in problems such as the leafminer mid-altitude and highland areas Liriomyza huidobrensis (Blanchard) and ranging from 700 to 2800 m above sea pod borers such as Epinotia aporema level and that Bemisia tabaci (Walsingham) on common bean and the (Gennadius), the tobacco whitefly, Neoleucinodes elegantalis (Gueneé) on ranges from sea level to 900 m above tomato. As shown in the previous sea level. Thus, these two species chapter, excessive insecticide use has overlap in many of the mid-altitude resulted also in the development of valleys. This is important because high levels of resistance of whiteflies to proper identification of whitefly species conventional insecticides. Some is a critical step in the formulation of farmers are now using costly novel research strategies. If, for example, insecticides that may become useless B. tabaci is erroneously assumed to be in a few years because of insecticide

296 Conclusions and Recommendations resistance. Knowledge on insecticide costly. In the long-term, the need to use patterns and insecticide resistance use more expensive insecticides may levels obtained in this first, diagnostic, threaten the economic viability of phase of the TWF-IPM Project will be common bean and tomato production important in the formulation of in mid-altitude and highland areas insecticide resistance management where small-scale, usually poor, strategies and in the development of farmers predominate. IPM options aimed at reducing pesticide use. Recommendations Insecticide resistance There is an urgent need to develop and In highland and mid-altitude areas of implement IPM systems that will help Colombia and Ecuador, T. vaporariorum re-establish the ecological equilibrium. showed high resistance to Experience in the Sumapaz region of methamidophos (an organophosphate), Colombia has shown that this is intermediate resistance to feasible: implementation of a relatively cypermethrin (a pyrethroid) and low simple IPM package based on cultural resistance to methomyl (a carbamate), control and sanitation practices, timely three test products chosen as application of effective insecticides and exemplifying the major groups of reliance on natural control in certain conventional insecticides currently in areas resulted in a 66% reduction in use. In general, resistance levels seem insecticide use (Cardona, 1995). to be related to intensity of insecticide Although the system was economically use. Responses of T. vaporariorum to and technically viable, adoption rates carbofuran and imidacloprid were varied with geographical area within determined, to provide a baseline for the region and were not as high as future monitoring of resistance against researchers expected. A main obstacle these products, both of which are in encountered was the high level of risk widespread use. The B biotype of aversion (as defined by Tisdell, 1986) B. tabaci, recently introduced into among small-scale farmers. However, Colombia, showed high levels of overall adoption rates of 30%-35% resistance to methomyl and were regarded as high by social methamidophos and, in some places, scientists linked to the project (Nohra moderate levels of resistance to Ruiz de Londoño [CIAT Impact Unit], cypermethrin. personal communication, 1996) and suggest that careful planning, active Development of resistance to participation of farmers in decision representative conventional making, and participatory research insecticides was higher than those approaches may help overcome reported in a previous study (Buitrago barriers to adoption. et al., 1994). The diagnostic survey showed that insecticide resistance is Several promising IPM tactics have widespread, affecting common bean been identified (Cardona, 1995; CIAT, and tomato growers alike. These are 2000). These include the replacement important findings that help explain of broad-spectrum insecticides with why conventional insecticides are no more selective ones, and the timing of longer effective and why farmers are applications according to pre- now using excessive numbers of established action thresholds. In applications of conventional pesticides addition, cultural control measures and new generation pesticides that are such as the incorporation of crop

297 Whiteflies and Whitefly-borne Viruses in the Tropics residues, leaf roguing and Based on the diagnostic phase manipulation of planting dates may be survey work, IPM pilot studies should effective in some circumstances. be continued in the following hot spots: Advances have been made also in the Sumapaz, Oriente Antioqueño, evaluation of the parasitoid Amitus Tenerife, Pradera and Nariño in fuscipennis MacGown and Nebeker Colombia, and in the Chota valley in (Manzano et al., 2000) and the northern Ecuador. However, problems entomopathogen Verticillium lecanii associated with T. vaporariorum extend (Zimm.) (González and López-Avila, far beyond the Colombian and 1997) but more work needs to be done Ecuadorian highlands and common in order to measure the effectiveness bean. This whitefly also has been of these and other natural enemies. reported as a significant field pest These preliminary efforts to develop affecting melon (Cucumis melo L.) and and test IPM strategies locally, tomato in Mexico (Flores et al., 1995), coupled with findings from the tomato in Costa Rica (Alpizar, 1993), diagnostic phase of the TWF-IPM tomato and common bean in the Project, can be used as a basis for Dominican Republic (FUNDESA, 1996), developing more widely applicable IPM as well as tomato, common bean and strategies. The overall results of these snap bean in Venezuela (Arnal et al., efforts should be the development of 1993), Ecuador (Mendoza, 1996), systems that will: (1) reduce levels of Colombia (Cardona, 1995), Bolivia pesticide used by common bean (CIAT, unpublished records), Brazil farmers; (2) reduce health risks to (Gerk et al., 1995), Peru (Núñez 1995) farmers and consumers; and and Argentina (L’Argentier et al., 1996; (3) re-establish the ecological Giganti et al., 1997). T. vaporariorum is equilibrium in areas that have been also a major pest in greenhouses in disturbed by excessive pesticide use. Colombia (Corredor et al., 1999) and Argentina (Viscarret and Botto, 1996; The immediate objective of any Salas et al., 1999). The benefit of future IPM initiative should be to developing effective IPM approaches for reduce pesticide use. Farmers T. vaporariorum is therefore likely to be currently depend so heavily on felt beyond the highlands of Colombia chemical pesticides, and and Ecuador. environmental disruption is of such magnitude, that for the foreseeable In order to overcome existing future it is most unlikely that any IPM barriers to adoption (Cardona, 1995), system could be effective without a IPM components should be tested with chemical component. In the longer local communities using participatory term, as more farmers adopt IPM research methods (Ashby, 1990). A approaches and environmental possible approach that already has equilibrium is re-established, more shown promise in Latin America ambitious, pesticide-free, strategies (Brown and Hocdé, 2000) is to involve might become feasible; however, for Comités de Investigación Agrícola Local the moment at least, monitoring of (CIALs) in the testing (and, in some resistance should continue and sound cases, validation) of individual IPM strategies to manage insecticide components and of promising IPM resistance will have to be developed. strategies. If major changes in existing Testing of new chemicals that are practices are found to be necessary effective but less detrimental to the such as the general adoption of environment will also be important. particular cropping dates or crop

298 Conclusions and Recommendations

rotations, then the implementation of Ashby, J. A. 1990. Evaluating technology IPM will have to be discussed with local with farmers. A handbook. Centro communities and government. The Internacional de Agricultura Tropical (CIAT), Cali, CO. 95 p. success of IPM approaches will depend ultimately on establishing consensus Brown, A. R.; Hocdé, H. 2000. Farmer among IPM stakeholders. Given participatory research in Latin farmers’ attitudes to chemical America: Four cases. In: Working pesticides and their current reliance on with farmers: The key to adoption of these products as the sole method of forage technologies. Procs. addressing whitefly problems, farmer International Workshop, 12-15 October 1999, Cagayan de education (Whitaker, 1993) will be Oro City, Mindanao, PH. Australian most important; however, if a major Centre for International Agricultural shift towards IPM approaches is to be Research (ACIAR), Canberra, AU. achieved, advocacy efforts directed at p. 32-53. consumers and the various stakeholders involved in agricultural Buitrago, N. A.; Cardona, C.; Acosta, A. production and processing is likely to 1994. Niveles de resistencia a insecticidas en Trialeurodes be necessary as well. vaporariorum (Westwood) (Homoptera: Aleyrodidae), plaga del Whitefly problems in the Andean fríjol común. Rev. Colomb. Entomol. highlands are evidently serious, 20(2):109-114. extensive and deep-rooted, so immediate success cannot be Cardona, C. 1995. Manejo de Trialeurodes vaporariorum anticipated. However, a phased (Westwood) en frijol en la zona Andina: Aspectos técnicos, approach, as conceived within the actitudes del agricultor y framework of the TWF-IPM Project, transferencia de tecnología. promises tremendous benefits to Memoria IV Taller Latinoamericano producers, consumers and the sobre moscas blancas y environment. The necessary geminivirus, 16-18 octubre 1995, investment in research, education and Zamorano, HN. CEIBA 36(1):53-64. policy change should therefore be Cardona, C.; Rodríguez, A.; Prada, P. C. undertaken without delay. 1993. Umbral de acción para el control de la mosca blanca de los invernaderos, Trialeurodes References vaporariorum (Westwood) (Homoptera: Aleyrodidae), en Alpizar, M. D. 1993. Aspectos básicos habichuela. Rev. Colomb. Entomol. sobre las moscas blancas con 19(1):27-33. énfasis en Bemisia tabaci y Trialeurodes vaporariorum. Bol. CIAT (Centro Internacional de Agricultura Divulgativo no. 112, Ministerio de Tropical). 2000. Annual report, Agricultura y Ganadería, San José, Project IP-1. Bean Improvement for CR. 22 p. Sustainable Productivity, Input Use Efficiency, and Poverty Alleviation. Arnal, E.; Russell, L. M.; Debrot, E.; Cali, CO. p. 143-147. Ramos, F.; Cermeli, M.; Marcano, R.; Montagne A. 1993. Lista de Corredor, D.; Fischer, G.; Angarita, A. moscas blancas (Homoptera: 1999. Integrated pest management Aleyrodidae) y sus plantas in cut flower crops grown in plastic hospederas en Venezuela. Fla. houses at the Bogotá Plateau. Entomol. 76(2):365-381. Procs. Int. Symp. on cut flowers in the tropics, 14-17 October 1997, Bogotá, CO. Acta Horticulturae 482:241-246.

299 Whiteflies and Whitefly-borne Viruses in the Tropics

Flores, A.; González, M. C.; Alarcón, A.; Manzano, M. R.; van Lenteren, J. C.; Ferrera R. 1995. Bioproducción de Cardona, C.; Drost, Y. C. 2000. melón en campo. Rev. Chapingo Developmental time, sex ratio, and Serie Hortic. 1(4):83-88. longevity of Amitus fuscipennis MacGown and Nebeker FUNDESA (Fundación para el Desarrollo (Hymenoptera: Platygasteridae) on Agropecuario, Santo Domingo). the greenhouse whitefly. Biol. 1996. Proyecto de evaluación de Control 18(2):94-100. variedades e híbridos de tomate por tolerancia a geminivirosis Mendoza, J. 1996. Qué está pasando con transmitida por moscas blancas. la mosca blanca en el Ecuador? Boletín FDA no. 9:2, 8. Santo Rev. INIAP 8: 8-10. Domingo, DO. Núñez, E. Y. 1995. Reporte de Perú. Gerk, A. O.; Vilela, E. F.; Pires, C. S.; Memoria IV Taller Latinoamericano Eiras, A. E. 1995. Biometría e ciclo sobre moscas blancas y de vida da mosca branca, geminivirus, 16-18 octubre 1995, Trialeurodes vaporariorum Zamorano, HN. CEIBA 36(1): (Westwood) e aspectos da 157-162. orientacao de seu parasitoide Encarsia formosa Gahan. An. Soc. Salas, H.; Fernández, M.; Willink, E. Entomol. Bras. 24(1):167-171. 1999. La mosca blanca de los invernáculos en cultivos hortícolas Giganti, H.; Dapoto, G.; González, J. R.; de Tucumán. Avance Agroind. Junyent, R. 1997. Insectos y ácaros (Argentina) 19:39-41. asociados a los principales cultivos hortícolas del Alto Valle del Río Tisdell, C. 1986. Levels of pest control Negro y Neuquén, Argentina. Hortic. and uncertainty of benefits. Aust. J. Argentina 16:40-41. Agric. Econ. 30:157-161.

González, J. G.; López-Avila, A. 1997. Viscarret, M. M.; Botto, E. N. 1996. Evaluación de cepas nativas de Descripción e identificación de Verticillium lecanii (Zimm.) en el Trialeurodes vaporariorum control de la mosca blanca de los (Westwood) y Bemisia tabaci invernaderos Trialeurodes (Gennadius) (Hemiptera, vaporariorum (Westwood). Rev. Homoptera: Aleyrodidae). Rev. Chil. Colomb. Entomol. 23(1-2):25-30. Ent. 23:51-58.

L’Argentier, S. M.; de Manero, E. A.; Whitaker, M .J. 1993. The challenge of Liacono A. 1996. Especies de pesticide education and training for Aleyrodidae (Homoptera) presentes tropical smallholders. Int. J. Pest en poroto (Phaseolus vulgaris L.) y Manag. 39(2):117-125. en las malezas asociadas al cultivo en las provincias de Jujuy y Salta, República Argentina. Rev. Investig. Centro de Investigaciones para la Regulación de Poblaciones de Organismos Vivos 10:1-4.

300 SECTION FIVE

Special Topics on Pest and Disease Management BLANCA 302 Sustainable IPM through Host Plant Resistance

CHAPTER 5.1 Sustainable Integrated Management of Whiteflies through Host Plant Resistance

Anthony Bellotti*, Joseph Tohme*, Michael Dunbier** and Gail Timmerman**

Introduction Research on whitefly control in the neotropics has emphasized host As direct feeding pests and vectors of plant resistant (HPR) and biological plant viruses, whiteflies constitute a control. The bulk of this report will major problem in the production of cover HPR research on whiteflies cassava (Manihot esculenta Crantz) in (Aleurotrachelus socialis Bondar) as a Africa, the neotropics and, to a lesser component of the Cassava Whitefly degree, Asia. The largest complex in the Integrated Pest Management (IPM) neotropics includes Bemisia tabaci Project. (Gennadius)1, B. tuberculata (Bondar) and Trialeurodes variabilis In traditional production (Quaintance). B. tabaci has a systems, resource-limited farmers pantropical distribution, feeding on have few options available for cassava throughout Africa, where it controlling pests. The cassava transmits African Cassava Mosaic germplasm bank at the Centro Disease, and several countries in Asia, Internacional de Agricultura Tropical including India and Malaysia. (CIAT) has nearly 6000 accessions and locally selected cultivars (land Whitefly feeding affects cassava in races) collected primarily in the three ways. Direct damage is caused by neotropics. These traditional cultivars feeding in the phloem of the leaves, represent centuries of cassava inducing chlorosis and leaf fall, which cultivation in diverse habitats, having results in considerable reduction in been selected by farmers over a long root yield if prolonged feeding occurs. period in the presence of a high Yield losses of this type are common in diversity of herbivores. These land the neotropics. Whiteflies also produce races often possess traits that confer a honeydew, which provides a medium low to moderate levels of resistance to for sooty mould growth that can reduce multiple pests. This germplasm bank yields. Most importantly, B. tabaci is a is constantly being evaluated for major vector of cassava viruses. resistance to several arthropod pests that can cause yield losses in cassava. Evaluations often are done

* Centro Internacional de Agricultura Tropical in more than one ecosystem. (CIAT), Cali, Colombia. ** Crop and Food Research, Levin, New The general purpose of this Zealand. project is “to reduce crop losses due 1. Evidence suggests that Bemisia tabaci represents a species complex with numerous to whitefly feeding damage and biotypes. whitefly-transmitted viruses, and

303 Whiteflies and Whitefly-borne Viruses in the Tropics prevent further environmental Cassava Germplasm degradation and food contamination Evaluation due to excessive pesticide use, leading to more productive and sustainable Field screening of cassava germplasm agricultural systems”. can be done at several sites in Colombia. Ideally, field populations of The project has three major whiteflies should be high and damage objectives, which are to: levels significant so as to distinguish susceptible cultivars. Field evaluations (1) Identify and access exotic or novel of cassava germplasm use a population genes and gene combinations that (nymphs) scale combined with a leaf can contribute to germplasm damage scale (Table 1). Evaluations are enhancement for whitefly made periodically throughout the resistance in cassava; growing cycle; four to five evaluations (2) Study the genetics of resistance are done, 1.5 to 2.0 months apart. and to map genes for whitefly During the 1998 growing cycle, cassava resistance in cassava and develop clones were evaluated at CIAT in the molecular markers for their Cauca Valley and Nataima in the incorporation into improved Tolima valley of Colombia. African, Latin American and Asian germplasm; and From 1994 through to 1996, (3) Develop crop management options whitefly (A. socialis) populations at for reducing whitefly populations Nataima, Tolima, were lower than and the transmission of whitefly- normal. Low whitefly populations transmitted viruses. provide inadequate selection pressure to ensure reliable resistance The research presented in this evaluation. All evaluations done at chapter is being funded by the Ministry Nataima are with natural field of Foreign Affairs and Trade (MFAT) populations of whiteflies. Host Plant Resistance (HPR) Project and consists of four major areas of During 1997-98, whitefly activity: populations increased significantly, allowing for more accurate germplasm (1) Cassava germplasm evaluation to evaluation. During 1999, some identify sources of whitefly 1651 cassava clones were planted in resistance in land race varieties in observational fields, and 1418 were the CIAT cassava germplasm evaluated. These clones were the F1 collection; progeny from crosses of cassava clones (2) Identification of genomic regions previously selected for their resistance responsible for the determination to whiteflies, with clones with desirable of whitefly resistance in cassava; agronomic characteristics. These (3) Identification of resistance crosses resulted in 17 families mechanisms in cassava; and (Table 1). Four field evaluations were (4) Tritrophic relationships to done over a period of several months, determine the effect of HPR on using a whitefly damage and whitefly parasitism. populations scale, as previously described.

Results indicate that whitefly populations were high and selection

304 Sustainable IPM through Host Plant Resistance

Table 1. Families of cassava clones formed from crosses of whitefly (Aleurotrachelus socialis)-resistant and -susceptible cultivars.

Family Crossesa Observation Female Male

CM 3317 MBra 12 (T) MCol 1468 (S) MCol 1468 = CMC-40 CM 5438 MBra 12 (T) MCol 1505 (S) CM 7559 MNGua 2 MBra 12 (T) CM 8884 CG 489-4 (R) MCol 1468 (S) CG 489-4 = MEcu 72 X MBra 12 CM 8885 CG 489-4 (R) MCol 1505 (S) CG 489-4 = MEcu 72 X MBra 12 CM 8887 CG 489-4 (R) MCol 2256 CG 489-4 = MEcu 72 X MBra 12 CM 8889 CG 489-23 (R) MCol 1468 (S) CG 489-23 = MEcu 72 X MBra 12 CM 8891 MCol 1468 (S) CG 489-34 (R) CG 489-34 = MEcu 72 X MBra 12 CM 8892 MCol 2246 CG 489-34 (R) CG 489-34 = MEcu 72 X MBra 12 CM 8893 MCol 2256 CG 489-34 (R) CG 489-34 = MEcu 72 X MBra 12 CM 8960 MCol 2246 MBra 12 (T) CM 8961 MCol 2256 MBra 12 (T) CM 8984 MCol 1505 (S) CG 489-34 (R) CG 489-34 = MEcu 72 X MBra 12 CM 8990 MCol 2026 CG 489-34 (R) CG 489-34 = MEcu 72 X MBra 12 CM 8991 MCol 2026 MBra 12 (T) CM 8995 MEcu 72 (R) MCol 1468 (S) CM 8996 MEcu 72 (R) MCol 2246 a. R, whitefly resistant cultivar; T, whitefly tolerant cultivar; and S, whitefly susceptible cultivar.

pressure was good. Of the 1418 clones evaluated, 938 (66%) had 800 damage ratings above 3.6 and were 703 eliminated as susceptible (Figure 1); 700 308 (22%) clones had an intermediate damage evaluation (2.6 to 3.5). The 600 remaining 172 (12%) clones had damage ratings below 2.6 and are 500 considered promising for resistance; 56 (3.9%) clones had damage ratings 400 of 1 to 1.5, indicating possible high

No. clones 308 levels of resistance. However, one 300 cycle (year) of field evaluation is not 235 adequate to identify resistance 200 confidently and these clones will continue to be evaluated for at least 116 100 3 cycles (years). A 1.0 to 1.5 rating 56 signifies no damage symptoms and 0 low whitefly population levels. The 1.5 2.5 3.5 4.5 6 families that most frequently occurred Damage grade in this group were CM 8984, with Resistant Intermediate Susceptible nine selections, and CM 8893, with six. In both these families, the Figure 1. Evaluations of F1 cassava clones resistant parent was CG 489-34, from crosses of whitefly (Aleurotrachelus socialis)-resistant indicating that this clone may contain and -susceptible cultivars at the good heritable resistant traits. Corporación Colombiana de Investigación Agropecuaria (CORPOICA), Nataima, Tolima (1997-98).

305 Whiteflies and Whitefly-borne Viruses in the Tropics

Whitefly populations at CIAT were populations. These two varieties will moderate to high for the fourth continue to be evaluated at CIAT and consecutive year. Evaluations at CIAT Tolima and should enter into breeding during 1998 concentrated on cultivars schemes for improved whitefly- that had been selected as promising for resistant clones. resistance during previous years at the Instituto Colombiano Agropecuário (ICA)-Nataima. Identification of Genomic Regions Responsible for Thirty-two clones sown in 50-plant the Determination of plots were evaluated using a 1 to 6 (low Whitefly Resistance in to severe damage or low to high Cassava whitefly population level) scale. Whitefly population scales are based on Previous research at CIAT determined counts of nymphs, pupae and adults. the existence of different sources of The 32 clones evaluated consisted of resistance to whitefly. The most land race varieties, hybrids and important genotypes are MBra 12 and backcrosses, and previously had shown MEcu 72, which were used as good levels of resistance in screening parentals in the generation of new trials in Tolima. Five regional or farmer genotypes. CG489-34 has shown the varieties from the Tolima area also were best resistance behaviour. The more included. susceptible genotypes, MCol 2026 and MCol 1505, also were selected and Results showed that nine cultivars, evaluated. or 26.5%, presented very low damage levels (1.0 to 1.5) and three cultivars The goal of this project is to study had damage levels between 1.6 and and screen plants with resistance to 2.5. The remaining 20 cultivars had whitefly in cassava, using molecular damage levels between 2.6 and 5. The techniques. regional cultivars from Tolima, Azucena, Ceiba Blanca, Almidona, Different breeding populations Llanera Precoz and Cuero de Marrano were obtained from the resistant and had damage and population ratings susceptible genotypes as parentals: between 4 and 4.5, indicating that farmer varieties in the regions are CG 489-34 X MCol 2026 = susceptible to whiteflies and probably 131 individuals experiencing significant yield losses. CG 489-34 X MCol 1505 = 108 individuals The nine best cultivars were MBra 12 X MCol 2026 = MEcu 64, MPer 335, MPer 415 and 135 individuals MEcu 72 (all land race varieties), and CM 8424-6, CM 8424-33, CM 8424-4, We selected the more contrasting CG 489-34 and CG 489-4 (all hybrids). individuals (resistant and susceptible) All had damage ratings of 1.0, except in the field for each family. DNA was MPer 415 and CG 489-34 with ratings extracted from the different individuals of 1.5. MEcu 72 and CG 489-34 had of each group and, with the parental, maintained low damage ratings was mixed in a bulk. We used consistently, over several years. The molecular screening (DNA restriction varieties MEcu 64 and MPer 335 had fragment length polymorphisms excellent growth habits as well as low [RFLPs] and random amplified damage ratings and low whitefly polymorphic DNA [RAPDs]) intended to

306 Sustainable IPM through Host Plant Resistance find markers associated to resistance, established with those reported on and later it is intended to isolate the other crops, to understand expression responsible genes. The parental lines patterns. The RFLP, RAPD, and the bulk will be screened with microsatellites and AFLP markers will amplified fragment length be used to generate a framework map polymorphism (AFLP) markers. and for the quantitative trait locus (QTL) analysis. Parentals from each family were evaluated with RFLP markers from the cassava map (Fregene et al., 1997). Identification of Seventy-two polymorphic probes were Resistance Mechanisms obtained. Sixty-two RAPD markers have been screened with the bulks of the same families; the primer OP.P3 showed Whitefly feeding behaviour a clear polymorphism between the susceptible and resistant group. This Cassava genotypes with resistance to marker is being evaluated with the whiteflies have been identified at CIAT. whole population of each cross to Resistant clone MEcu 72 shows high confirm its association with the mortality for both adult and immature resistance (Figure 2). whiteflies, which may suggest less feeding on this genotype under natural We intend to isolate and sequence conditions. It is necessary to identify the polymorphic bands and generate a whitefly feeding behaviour on sequence characterized amplified region susceptible genotypes and to make (SCAR). This can be used to make a comparisons with MEcu 72 in order to diagnosis of resistant materials and better understand the mechanisms of determine the most promising resistance. genotypes for breeding programs and farmers. Electronic monitoring of insect feeding (EMIF) is a technique that With the sequence of the genes of permits the identification and resistance, homologies will be quantification of the feeding behaviours

1 1 2 3 3 4 R and S : CG 489-34 x MCol 2026 R and R and S : CG 489-34 x MCol 1505 R : S2: MBra 12 x MCol 2026 Ecu 72; R5: Ecu 72 progeny

Figure 2. Random amplified polymorphic DNA (RAPD) bulk analysis of whitefly (Aleurotrachelus socialis)-resistant and -susceptible cassava clones.

307 Whiteflies and Whitefly-borne Viruses in the Tropics of hemipteran insects. By passing an oxidizer that dissolves away part of the electrical signal to a test plant, and gold, leaving a thinner wire while tethering an insect with a fine gold conserving its electricity conducting wire, modifications caused by stylet properties. Pieces of this thin gold wire movements and feeding behaviours can are attached with silver paint to a be observed as waveforms. The EMIF copper stub. technique has been used extensively for the study of mechanisms of plant Female whiteflies from a resistance to insects. CIAT presently greenhouse colony are placed by mouth owns two AC-Electronic feeding aspirator on a vacuum stand and held monitors, and has access to a DC in place by a gentle vacuum. Several version of the system (electrical individuals can be placed penetration graph [EPG]). The AC simultaneously. With the help of the systems have been used with copper stub, the thin gold wire is leafhoppers (Empoasca kraemeri Ross & placed on the mesonotum of a female, Moore) on common bean (Phaseolus and a small drop of electrically vulgaris L.), and the DC systems with conducting silver paint is used to cassava mealybugs (Phenacoccus attach the wire to the insect. Care manihoti Matile-Ferrero). Preliminary needs to be taken to not get silver paint observations with both systems on the whitefly’s eyes or wings. This suggested that an easy protocol could affects their behaviour even more than be developed for monitoring the the tethered condition. New insects whitefly. In addition, CIAT technicians need to be used should silver paint get needed to be trained in wiring on their wings or eyes. techniques, operation of the system, computer display and data acquisition. Tethered whiteflies can be acclimated to the wire by placing them Electronic monitoring on a cassava leaf for 1 hour. Before the methodology electronic monitoring session begins, whiteflies are starved for 30 minutes. The methodology devised for electronic The copper stub holding the tethered monitoring of whitefly feeding whitefly is attached to the alligator clip behaviour includes two major steps: a on one of the input electrodes of the wiring technique to attach the thin wire electronic monitor. to the mesonotum of an adult whitefly and the proper settings (ground A constant signal, 250 mV at voltages, signal frequency) of the 250 Hz, was established as the best electronic monitoring system. setting for A. socialis. This signal is transmitted via an electrode inserted in The “normal” gold wire used with the substrate of a potted plant (4-8 leafhoppers (12.7 µm) is too thick and weeks old). Detached leaves placed on stiff for the small whitefly adults. A a container filled with water and sealed thinner wire needed to be obtained. with the output electrode of the Since purchase of a thinner wire was electronic monitor also worked and difficult in the short period of time showed better conductivity than the allowed for this project, thinning of the potted plants. However, the effects of existing gold wire was necessary. To do detached leaves vs. whole plants on this, a 10- to 20-cm piece of the thick whitefly feeding behaviour need to be wire is placed for 45 min in a solution evaluated further. of 3 mol nitric acid and 9 mol hydrochloric acid. This is a potent

308 Sustainable IPM through Host Plant Resistance

Feeding behaviour that waveform patterns are very similar During probing, stylet movements and among them. Since waveforms shown are similar in appearance to those other feeding behaviours induce changes to the constant signal. These reported in the literature, we will modifications, known as waveforms, describe them. All probes observed started with salivation waveforms are captured in a computer using an analogue to digital converter board, corresponding to the intercellular path displayed on the screen on a time that the stylets follow in their way to the phloem. These waveforms are scale, and stored for post-acquisition measurement and analysis. Once shown in Figures 3A and 3C as waveforms have been correlated with “intercellular stylet path”. They have been correlated also with the specific behaviours, quantification can be made on their frequency and deposition of a salivary sheath. In some duration. instances, the stylets come across treachery elements in the xylem and the whitefly ingests from them. This Waveforms produced by probing particular waveform has not been A. socialis on CMC 40 are shown in correlated with completely, but there is Figure 3. Because of time limitations, evidence of ingestion. After reaching waveforms cannot be correlated with the phloem, normally whiteflies go into specific behaviours for this whitefly continued ingestion, producing a flat species. However, it has been reported waveform whose beginning is that several species of whiteflies share illustrated toward the end of the trace very stereotypical styles of feeding and (Figure 3).

A Intercellular Stylet Path (Salivation)

Phloem B

C Intercellular Stylet Path (Salivation) Possib. Xylem ingestion

Unknown

Phloem D Intercellular Stylet Path (Salivation)

Figure 3. Waveforms produced by probing Aleurotrachelus socialis on cassava cv. CMC 40. A and B: Single probe (stylet penetration). B and C: Single probe forms a second whitefly. Note that waveform identified as possible xylem ingestion and an unknown waveform that has yet to be correlated with a feeding behaviour.

309 Whiteflies and Whitefly-borne Viruses in the Tropics

Figure 3D shows a new waveform Plants of the abovementioned that previously has not been identified varieties at 1 month of age were or associated with any behaviour. It infested with A. socialis eggs by would be expected that, with more exposing the leaf undersurface to electronic monitoring of whitefly ovipositing adults for 36 hours. After feeding, more patterns would be 15 days, when whitefly immatures were identified and more correlational work in the second instar, they were exposed would be needed. Now that a to E. hispida parasites. E. hispida was methodology exists for the study of obtained by collecting cassava leaves whitefly feeding behaviour at CIAT, with parasitized whitefly pupae in the comparisons can be made among field at CIAT, placing the leaves in resistant and susceptible genotypes. blackened plastic traps and capturing emerging parasites. The results at this point are preliminary. Tritrophic Relationships: Determining Effect of Results to date indicate that the HPR on Whitefly female longevity of E. hispida was Parasitism similar on varieties CMC 40 (17.8 days) and CG 489-34 in experiments 1 and It has been observed that farmers will 2 (12.2 and 13.8 days, respectively) use pesticides to control high (Figure 4). E. hispida longevity on populations of cassava whiteflies. MEcu 72 was greatly reduced, only Pesticide applications will reduce the 7.7 days, indicating that this variety effectiveness of biological control as has factors or characteristics that well as cause environmental affect the longevity of the parasite. contamination. Several years of These factors could be the number of research at CIAT has identified tricomes or a chemical component that cultivars with varying levels of is interfering with parasitoid feeding resistance to whiteflies, especially the (Table 2). species A. socialis.

Present research is investigating 20 17.8 13.83 the compatibility between HPR and 15 12.2 biological control. The combination of 10 7.77 these two methods could reduce pest 5 Time (days) populations below economic injury 0 levels or extend the usefulness of HPR. CMC CG 489- CG 489- MEcu 40 34 #1 34 #2 72 In addition, it is important to know the Variety effect that plant resistance, especially antibiosis, might have on parasitoid Figure 4. Longevity of the parasitoid Encarsia hispida on the whitefly populations. Aleurotrachelus socialis on three cassava varieties, CMC 40, The compatibility of the parasitoid CG 489-34 (experiments #1 and #2) and MEcu72. Encarsia hispida DeSantis was evaluated on whitefly immatures feeding on whitefly resistance E. hispida emergence from A. genotypes of cassava. The resistant socialis pupae was measured on the varieties tested were MEcu 72, and varieties CG 489-34 and MEcu 72. CG 489-4; the susceptible check was Results show a low rate of emergence CMC 40. Parasite development time, of the parasitoid, 0.43 for CG 489-34 fecundity and survival were measured. and 0.86 for MEcu 72. These

310 Sustainable IPM through Host Plant Resistance

Table 2. Whitefly (Aleurotrachelus socialis) survival, development time and fecundity on four cassava varieties.

Variety Percentage survival Development Number Tricomes/cm2 of immaturesa time/daysa eggs/female on second leaf

MBra 12 75.0 a 32.18 c 107 4.653 CMC 40 66.0 ab 32.07 c 64 189 CG 489-34 65.5 ab 33.13 b 46 15.290 MEcu 72 27.0 c 34.45 a 15 33.048 a. Means within a column followed by the same letter are not significantly different at the P = 0.05 level, using LSD test.

preliminary results indicate that host individuals survived until 34 days. plant resistance in cassava may affect These data indicate that there may be the development or emergence of varietal influence associated with parasitoids. Experiments using E. hispida survival. Since MEcu 72 and susceptible varieties (CMC 40) still CG 489-34 are resistant varieties, this need to be completed. Since E. hispida resistance factor or mechanism may be is the most frequently collected affecting parasitoid survival. parasitoid in cassava fields, a much higher emergence is expected. Host plant resistance and biological control agents increasingly The survival of E. hispida when are accepted as complementary pest- associated with A. socialis on the control tactics that can reduce the use cultivar MEcu 72 decreased rapidly of chemical pesticide, especially with when compared to the other cultivars. difficult-to-control pests such as One hundred percent mortality whiteflies. Cassava is most often grown occurred within 19 days (Figure 5). by resource-limited, small-scale Longest survival was on CMC 40, with farmers in the tropical and sub-tropical individuals living for 40 days. Variety regions of the world. HPR and CG 489-34 was intermediate; in both biological control offers a low-cost, experiments with this cultivar, practical, long-term solution for

100

40 % survival

0 1 3 5 7 9 111315171921232527293133353739 Time (days)

MEcu 72 CMC 40 CG 489-34 #1 CG 489-34 #2

Figure 5. Survival of Encarsia hispida, parasitoid of Aleurotrachelus socialis, on three varieties of cassava, MEcu 72, CMC 40 and CG 489-34 (experiments #1 and #2).

311 Whiteflies and Whitefly-borne Viruses in the Tropics maintaining lower whitefly populations, made to identify and use natural reducing crop losses, and limiting or enemies in an IPM context (see eliminating the need for applications of Chapter 5.2). costly insecticides. This is especially important in crops such as cassava, which has a long growing cycle (1 year Reference or more) that would require numerous pesticide applications to achieve Fregene, M.; Angel, F.; Gómez, R.; adequate whitefly control. Research on Rodríguez, F.; Chavarriaga, P.; Roca, W.; Tohme, J.; Bonierbale, M. cassava whitefly control at CIAT 1997. A molecular genetic map of initially emphasized HPR. More cassava (Manihot esculenta Crantz). recently, a concentrated effort is being Theor. Appl. Gen. 95:431-441.

312 Biological Control of Whiteflies by Indigenous Natural Enemies

CHAPTER 5.2 Biological Control of Whiteflies by Indigenous Natural Enemies for Major Food Crops in the Neotropics

Anthony Bellotti*, Jorge Peña**, Bernardo Arias*, José María Guerrero*, Harold Trujillo*, Claudia Holguín* and Adriana Ortega*

Introduction of African cassava mosaic disease (ACMD) in the Americas may be related Whiteflies as direct feeding pests and to the inability of its vector, B. tabaci, to virus vectors constitute a major colonize cassava (Costa and Russell, problem in cassava (Manihot esculenta 1975). Since the early 1990s, a new Crantz) and associated crops in Central biotype (B) of B. tabaci, considered by and South America and the Caribbean some to be a separate species (Bemisia region. There is a large complex in the argentifolii Bellows & Perring) (Perring et neotropics, where 11 species are al., 1993) has been found feeding on reported on cassava alone (Bellotti et cassava in the neotropics (França et al., al., 1999). In the northern region of 1996; Quintero et al., 1998). More South America (Colombia, Venezuela, recently, in greenhouse studies done at and Ecuador) the major species is the Centro Internacional de Agricultura Aleurotrachelus socialis (Bondar) Tropical (CIAT), it has been shown that (Castillo, 1996). Two additional species, B. tabaci females will oviposit on of lesser importance, but frequently cassava, and immatures will feed and found on cassava, are Bemisia pupate; however, few adults emerged tuberculata (Bondar) and Trialeurodes (CIAT, 1999). It is considered that variabilis (Quaintance). High ACMD now poses a more serious threat populations of A. socialis, frequently to cassava production, because most observed in the region, can cause traditional varieties in the neotropics serious yield reductions in cassava. are highly susceptible to the disease. In There is a correlation between duration addition, the B biotype of B. tabaci as a of attack and yield loss: infestations of virus vector causes heavy crop losses 1 month resulted in a 5% yield on numerous other crops in the reduction, of 6 months in a 42% neotropics and these often are grown in reduction, and of 11 months in a 79% association with cassava, or in the reduction (Vargas and Bellotti, 1981). same area. The possibility of virus diseases moving between these crops, Until recently, the Bemisia tabaci or the appearance of previously (Gennadius) biotypes found in the unrecorded viruses, has become a Americas did not feed on cassava. It potential threat. has been speculated that the absence Host plant resistance and biological control agents (e.g., parasitoids, predators and entomopathogens) * Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. increasingly are accepted as a means of ** University of Florida, USA. pest control that reduces environmental

313 Whiteflies and Whitefly-borne Viruses in the Tropics contamination and other disadvantages A. socialis and B. tuberculata had a arising from excessive use of chemical wide distribution, both being collected pesticides. Many natural enemies are in almost all sites surveyed. found associated with species of T. variabilis was collected from the whiteflies on cassava in the neotropics, Atlantic Coast and the Andean region, especially in Colombia and Venezuela but not from the Eastern Plains. (Gerling, 1986; Gold et al., 1989; A. socialis populations were highest in Castillo, 1996; Evans and Castillo, the Eastern Plains and lowest in the 1998). Although a large complex of Andean region. B. tuberculata numbers parasitoids have been collected from were highest on the Atlantic Coast and cassava whiteflies, little is known about Eastern Plains, and almost all the levels of parasitism they impose on T. variabilis were collected from the the whitefly population, either as Andean region. The data from this individual species or collectively. study showed that A. socialis and B. tuberculata populations decreased For instance, during 1994 and with increasing altitude, and were most 1995, CIAT personnel carried out abundant below 900 m. T. variabilis surveys for cassava whitefly natural was most abundant at higher enemies in three regions of Colombia: elevations, above 1400 m. A. socialis the Andes, the Atlantic Coast and the populations were highest when Eastern Plains (Castillo, 1996). One temperatures were above 23 °C, while hundred sites were visited in T. variabilis populations were higher at 20 departments (states). The three lower temperature (<19 °C). regions differ in altitude, temperature B. tuberculata were most abundant in and precipitation, as well as in their those areas with higher temperatures. cassava cultivation patterns. The Annual precipitation did not have a Andean region contains many small significant effect on distribution of any plantations dispersed over a wide area; one of the three species. in the Atlantic Coast, cassava cultivation is generally in extensive and A species richness of whitefly uniform areas; while in the Eastern parasitoids was collected during these Plains, cassava is cultivated in small surveys. Ten species were collected “patchy” areas near the foot of the from A. socialis, B. tuberculata and easternmost range of the Andes. T. variabilis. Several of these were unrecorded species (Castillo, 1996; The whitefly species collected on Evans and Castillo, 1998). The cassava during these surveys were parasitoids collected were represented A. socialis, B. tuberculata, T. variabilis, by three genera, Encarsia (four Tetraleurodes sp. and Aleuroglandulus species), Eretmocerus (four species) malangae Russell. The latter two and Amitus (two species). Three of the species were collected for the first time Encarsia species were E. hispida in very low numbers in the Atlantic DeSantis, E. pergandiella Howard and Coast and Andean region. A. socialis E. bellotti. None of the Eretmocerus represented 64.5% of the adults and genus was identified to species and one 86% of the immature (eggs, nymphs, Amitus was identified to species, and pupae) collected across the three A. macgowni Evans and Castillo. regions. T. variabilis was represented Eretmocerus sp. “b”, E. hispida, by 27.5% of the adults and 11.0% of A. macgowni and E. bellotti were the immatures, while B. tuberculata collected parasitizing A. socialis. represented 8.0% of the adults and E. bellotti, E. hispida, E. pergandiella 3.0% of the immatures (Castillo, 1996). and Eretmocerus sp. “a” were collected

314 Biological Control of Whiteflies by Indigenous Natural Enemies from T. variabilis. Only Eretmocerus sp. (4) Determine if the B. biotype of “c” was collected parasitizing B. tabaci is infesting cassava fields. B. tuberculata (Castillo, 1996).

The highest level of parasitism was Methods around 15% measured on A. socialis in the Eastern Plains; almost 14% was measured on B. tuberculata and 12% Survey methodology on T. variabilis, both in the Andean Three geographic areas were selected for region. E. hispida represented the exploration for whitefly species and their greatest number of individuals parasitoid natural enemies. These were collected, 1845 or about 64% of all the Atlantic Coast (the Departments of parasitoids collected. Eretmocerus sp. Atlántico, Córdoba, Bolívar and “b” was the second most commonly Magdalena), and the mid altitude collected (485 individuals), followed by Central Highlands (Departments of A. macgowni with 159 individuals. Cauca, Valle del Cauca, Caldas, E. hispida and Eretmocerus sp. “b” were Quindío and Risaralda) of Colombia. collected in all three regions. The The surveyed Ecuadorian regions were greatest complex of species was found the coastal area and the highlands in the Andean region, whereas only (Sierra). E. hispida and Eretmocerus sp. “b” were observed in the Coastal and Eastern The Colombian departments Plains regions. surveyed represented two distinct ecological zones. The Atlantic Coast, The results of this initial phase of which is hot and fairly dry, has the surveys indicated a rich complex of temperatures ranging from 27 to 36 °C parasitoids associated with cassava and an average relative humidity (RH) whiteflies, many of which are from 25% to 70%. The Andean region is unrecorded species that could prove much cooler, with temperatures ranging useful in biological control programs. It from 22 to 33 °C, and RH ranging from was further realized that an expanded 7% to 100%. The Atlantic Coast is also survey would prove beneficial and characterized by a 4- to 6-month dry increase our knowledge of the natural season. Dry periods in the Andean enemy complex associated with region are usually 2 to 3 months. whiteflies. Altitude was 0 to 200 m for the collection sites on the Atlantic Coast, A second phase to these studies and 25 to 1750 m in the Andean region. was undertaken with the objectives to: Several crop hosts, including (1) Further define the complex of cassava, common bean (Phaseolus natural enemies, especially vulgaris L.), tomato (Lycopersicon parasitoids, and quantify their esculentum Mill.), eggplant (Solanum association with whitefly species; melongena L.), cotton (Gossypium (2) Expand these surveys to include hirsutum L.), cucumber (Cucumis sativus other crops in addition to cassava; L. var. sativus) and snap bean (3) Determine which natural enemies (P. vulgaris) were sampled. From each regulate whitefly populations by collection site, 100 leaves were collected evaluating parasitoid species under randomly and 1 square inch leaf area controlled conditions (laboratory was examined to determine the whitefly studies with E. hispida, a parasite species present and their respective of A. socialis, were initiated); and densities.

315 Whiteflies and Whitefly-borne Viruses in the Tropics

The rate of parasitism was instar A. socialis nymphs; 16 to 18 days determined by collecting 40 leaves after release, adult parasitoids began to randomly from the field and removing a emerge. Once emergence was detected, 1-inch square from each leaf sampled. cassava leaves (now with parasitized Only one whitefly species was allowed to pupae) were collected and placed in the remain on each leaf square and the “emergence chamber” where parasitoids emergence of parasitoids is recorded for were collected and species identification each whitefly species. This methodology confirmed. Collected parasitoids were allowed us to accurately determine the used to maintain the parasite colony, or parasitoid species associated with each utilized in the various studies described whitefly species. Identifications remain in this report. pending for some whitefly and parasitoid species. E. hispida parasitism studies were carried out in the greenhouse (25 to Methodology for parasitism 32 °C; 60%-80% RH), using A. socialis studies with Encarsia hispida as host species. Three methodologies were evaluated to determine parasitism Field surveys carried out over several efficiency: years have shown that the parasitoid E. hispida is frequently found (1) Four nylon-mesh cages (50 x 50 x parasitizing cassava whiteflies. A colony 90 cm) were used, containing two of this species was established on A. potted cassava plants infested with socialis on cassava in the greenhouse, whitefly (A. socialis) nymphs. Sixty and activities were initiated to study E. parasitoids were released into each hispida biology and behaviour. cage with four repetitions for each instar. To ensure successful rearing of (2) Four whitefly-infested cassava whitefly parasitoids, a colony of the plants were used; four infested whitefly A. socialis was maintained on leaves were placed in a petri dish cassava (var. CMC-40) in the (100 x 15 mm), and 30 parasitoids greenhouse. Potted 5-week old cassava were released into each petri dish plants were infested weekly by exposing with 16 repetitions for each instar. them to A. socialis adults for oviposition (3) A muslin “bag” was placed over an for 48 hours. The parasitoid colony was entire cassava leaf and sealed at initiated by collecting cassava leaves the petiole. Each leaf was with parasitized A. socialis pupae from infested with 80 nymphs, and the field. These were placed in a black 20 parasitoids were released into plastic box (emergence chambers) with each bag with four repetitions per a clear glass bottle attached to an instar. opening in the lid where emerging parasites are drawn to the light and collected. The parasitoids were removed Results from the bottle with an aspirator and identified. Surveys on the Colombian Atlantic Coast and the departments of Cauca The desired parasitoid species and Valle del Cauca resulted in four (E. hispida) was collected in sufficient species of whiteflies being confirmed as numbers and released into nylon-mesh feeding on cassava: Aleurotrachelus cages (50 x 50 x 90 cm) in the socialis, Bemisia tuberculata, greenhouse (25 to 32 °C, 60%-80% RH). Trialeurodes sp. (probably T. variabilis) Each cage contained potted cassava and Tetraleurodes sp. (Table 1). plants infested with second to third A. socialis was the predominant species

316 Biological Control of Whiteflies by Indigenous Natural Enemies

Table 1. Whitefly species and their associated parasitoid complex collected on cassava from three geographical regions of Colombia.

Region Whitefly species Parasitoid species

Atlantic Coast Aleurotrachelus socialis Encarsia sp. Eretmocerus sp. Bemisia tuberculata Encarsia sp. Eretmocerus sp. Metaphycus sp. Trialeurodes sp. Encarsia sp. Eretmocerus sp. Tetraleurodes sp. Valle del Cauca A. socialis Encarsia sp. Eretmocerus sp. B. tuberculata Cauca A. socialis Encarsia bellotti Eretmocerus sp. Signiphora aleyrodis B. tuberculata Encarsia pergandiella Eretmocerus sp. Euderomphale sp. Signiphora aleyrodis Trialeurodes sp. Encarsia hispida Encarsia pergandiella Eretmocerus sp.

found in all three areas; however, its collected in either Cauca or Valle del population was considerably higher in Cauca. Valle del Cauca and lowest in Cauca (Figure 1). All four species were Pupal populations of A. socialis collected from the Atlantic Coast, ranged from an average of about 55 per whereas Tetraleurodes sp. was not square inch on leaf samples from Valle del Cauca to about 5 on the Atlantic Coast and 2 in Cauca (Figure 1). The data also showed that whenever there 6000 are high whitefly densities or populations, A. socialis is the 5000 predominant species regardless of geographic area. However, when 4000 densities are low at sampling sites, other species may predominate (Figure 2). On 3000 the Atlantic Coast, under low population densities, B. tuberculata and A. socialis 2000 had similar populations, whereas in Valle del Cauca, B. tuberculata

No. of pupae/100 samples 1000 predominated. In Cauca, Trialeurodes had the highest population.

0 Atlantic Cauca Valle del Numerous whitefly parasitoids were Coast Cauca collected from the three areas (Table 1). Aleurotrachelus socialis Trialeurodes sp. These parasitoids belong to the genera Bemisia tuberculata Tetraleurodes sp. Encarsia, Eretmocerus, Metaphycus and Figure 1. Whitefly population densities on Euderomphale, and indications are that cassava in three geographic zones of the collection contains several Colombia. unrecorded species. The hyperparasitoid

317 Whiteflies and Whitefly-borne Viruses in the Tropics

100 100

80 80

60 60

40 Parasitoids

No. of pupae/100 in sq. 20 20

0 0 Atlantic Cauca Valle del Atlantic Cauca Valle del Coast Cauca Coast Cauca Aleurotrachelus socialis Trialeurodes sp. Encarsia sp. Euderomphale sp. Bemisia tuberculata Tetraleurodes sp. Eretmocerus sp. Signiphora aleurodis Metaphycus sp. Not identified Figure 2. The whitefly species complex at collecting sites with low population Figure 3. Parasitoid species (%) collected from densities in three geographic zones whiteflies on cassava in three of Colombia. geographic zones of Colombia.

Signiphora aleyrodis was collected only A. socialis, the predominant parasitoid in Cauca. Encarsia was the genera most genus in the Atlantic Coast was frequently collected in Cauca and Valle Eretmocerus, while in Cauca and Valle del Cauca (Figure 3), with the highest del Cauca, Encarsia predominated percentage in the latter. On the Atlantic (Figure 4). In Valle del Cauca, 99.6% of Coast, Eretmocerus was the most the parasitism of A. socialis was by frequent genus. Metaphycus was only Encarsia and 0.4% by Eretmocerus. The collected on the Atlantic Coast and most numerous complex of parasitoids Euderomphale only from Valle del was found associated with B. tuberculata Cauca. (Table 1). Eretmocerus and Metaphycus were the predominant genera on the Levels of infestations were high for Atlantic Coast and Eretmocerus in Cauca. Aleurodicus sp., A. socialis, The hyperparasite S. aleyrodis was B. tuberculata and Tetraleurodes sp. for collected from both A. socialis and the coastal region of Ecuador. B. tuberculata pupae. T. vaporariorum, which is found infesting common bean, was found for Four species of parasitoids were the first time at high infestation levels collected from Ecuador. The complex of in cassava in the coastal and highlands parasitoids collected from whiteflies in regions of Ecuador. cassava in Ecuador has been identified only to the generic level (Greg Evans, The association between whitefly Mike Rose, personal communication, species and parasitoid species was 1999). In the coastal region, A. socialis evaluated also. Results indicate a was parasitized by Encarsia, followed by sizable parasitoid species complex Amitus and Eretmocerus; Aleurodicus was associated with each whitefly species parasitized by Euderomphale sp. and that this can be influenced by Tetraleurodes and Trialeurodes were geographic area. For example, with mostly parasitized by Eretmocerus sp.

318 Biological Control of Whiteflies by Indigenous Natural Enemies

The whitefly species complex 100 associated with cotton and legume and vegetable crops was distinct from that 80 described on cassava. Two whitefly species were collected, Bemisia tabaci and Trialeurodes vaporariorum. During 60 the period of field collections (January to June 1999), B. tabaci was the only 40 species collected on the Atlantic Coast Parasitoids (departments of Atlántico and Córdoba), while in the Andean region 20 (departments of Caldas, Quindío, Risaralda and Valle del Cauca), T. vaporariorum was the only species 0 Atlantic Cauca Valle del collected, with its highest populations Coast Cauca observed in Valle del Cauca and Not identified Encarsia bellotti Caldas. Encarsia sp. Eretmocerus sp. Signiphora aleyrodis B. tabaci was collected from Figure 4. Parasitoid species (%) collected from cotton, eggplant and tomato, while the whitefly Aleurotrachelus socialis T. vaporariorum was collected from in three geographic zones of common bean, snap bean, cucumber Colombia. and tomato. Populations of T. vaporariorum in general were much and B. tuberculata was parasitized by higher than B. tabaci and this was Encarsia sp. and Euderomphale. In the especially the case on common bean highland region, T. vaporariorum was and snap bean (Figure 5). These the only whitefly species collected, with results indicate that B. tabaci may be about 16 pupae per 2.54 cm2. The adapted more to the warmer dominant parasitoids were Encarsia temperatures of the coastal region, spp., representing 98% of the sample. whereas T. vaporariorum prefers the cooler temperatures of the Andean In Colombia, Trialeurodes sp. was region. most frequently parasitized by Encarsia. Two species of Encarsia were identified, E. hispida and 1200 E. pergandiella, and a third species 1000 remains to be identified. 800 600 Encarsia sp. and Eretmocerus spp. 400 were frequently collected from both 200 high and low whitefly populations. (100 in sq plugs) Whitefly population 0 Metaphycus sp. and Euderomphale sp. Bemisia tabaci Trialeurodes vaporariorum are observed only when whitefly populations are low. Parasitoids of the Cotton Common bean Cucumber genera Encarsia and Eretmocerus were Eggplant Snap bean Tomato collected from sea level to about Figure 5. Whitefly nymph pupae populations 2400 m. E. pergandiella was collected on several crops in the departments of Atlántico, Córdoba, Caldas, in high populations at all levels of Quindío, Risaralda and Valle del whitefly infestation. Cauca, Colombia.

319 Whiteflies and Whitefly-borne Viruses in the Tropics

Parasitoids from four genera, Encarsia parasites were collected Encarsia and Eretmocerus from all the plant hosts sampled and (Aphelinidae), Amitus (Platygastridae) had the overall highest densities among and Metaphycus (Encyrtidae) were all collected parasitoids (Figure 7). collected. These collections have been Amitus, although not collected from all sent to taxonomists and species hosts, also presented high populations, identifications are pending. Therefore, especially on common bean and snap in this report, they are referred to only bean. Eretmocerus was associated by genera. almost exclusively with cotton (only one individual was collected from tomato). Encarsia was collected in all the departments except Risaralda (where only two sites were surveyed) but 70 Amitus presented the highest populations, especially in Valle del 60 Cauca and Caldas (Figure 6). This 50 corresponded to the two departments with the highest whitefly populations. 40 Eretmocerus was the predominant species in Córdoba. Parasitoid 30 populations were highest in Valle del 20

Cauca and lowest in Risaralda and Parasitoids/40 plugs Quindío. Only one specimen of 10 Metaphycus was collected across the 0 sites. bean bean Snap Cotton Tomato Eggplant Common Cucumber 70 Encarsia sp. Eretmocerus Amitus 60 Metaphycus Signiphora Figure 7. Whitefly parasitoid frequency 50 collected from several crops in the departments of Atlántico, Córdoba, 40 Caldas, Quindío, Risaralda and Valle del Cauca, Colombia. 30

20 Encarsia and Eretmocerus were Parasitoids/40 plugs collected from both whitefly species 10 (Figure 8), while Amitus was collected 0 only on T. vaporariorum and Metaphycus and Signiphora were Valle

Caldas collected only from B. tabaci. Quindío Córdoba Atlántico Risaralda Amitus Encarsia Eretmocerus During surveys, data on pesticide Metaphycus Signiphora applications were recorded during Figure 6. Whitefly parasitoids collected from farmer interviews. In general, several crops in the departments of parasitoid populations were higher in Atlántico, Córdoba, Caldas, Quindío, fields where pesticides were not Risaralda and Valle del Cauca, Colombia. applied, regardless of the whitefly species. Results show that populations of Encarsia are less affected by pesticide applications than populations

320 Biological Control of Whiteflies by Indigenous Natural Enemies

with about 75% parasitism in the third 60 instar and 19% in the 1st instar, 61% in the 2nd instar and 25% in the 50 4th instar (Figure 9). The average parasitism rate for these two 40 experiments were about 45%, whereas

30 average parasitism rates were only 6% using methodology 1 and were 20% 20 using methodology 2. Parasitoids (40 plugs) 10 80 0 70 Bemisia Trialeurodes 60 tabaci vaporariorum 50 Amitus Encarsia Eretmocerus 40 Metaphycus Signiphora 30 20 % parasitism Figure 8. Relationship between parasitoid 10 complex and whitefly species 0 collected from several crops in the I II III IV departments of Atlántico, Córdoba, Instars Caldas, Quindío, Risaralda and Exp. 1 Exp. 2 Valle del Cauca, Colombia. Figure 9. Cassava whitefly (Aleurotrachelus socialis) instar preference by the of Amitus, indicating that Encarsia may parasitoid Encarsia hispida in two greenhouse experiments. have acquired a degree of resistance to some of the pesticides applied. These results also show that the Parasitism studies with third whitefly instar is preferred for E. hispida parasitism by E. hispida. An average of Of the three methodologies evaluated, all four experiments resulted in a the highest rate of parasitism was parasitism rate of 21% for the obtained using the third method. 1st instar, 35% for the 2nd instar, 46% Parasitism levels using the first two for the 3rd instar and 22% for the methods were low. The highest rate of 4th instar. However, the average of the parasitism with method 1, the use of two experiments using methodology 3 nylon-mesh cages, only reached 9% in (muslin bags) is 17% for the 1st instar, the third instar. This improved with the 53% for the 2nd, 75% for the 3rd and second method (the use of leaves 34% for the 4th instar. The highest enclosed in petri dishes) to 30% in the parasitism rate is in the 3rd instar, 2nd instar and to 20% in the 3rd. followed by the 2nd, 4th and 1st instar.

The employment of muslin bags, as The time period for optimal described in the third methodology, parasitoid activity was evaluated using gave the best results. In the first of two the third methodology described. experiments using this methodology, Percent parasitism evaluations were parasitism rates reached about 75% in made at 48, 72, 96 and 216 hours after the third instar and 16% in the parasitoid release. Peak parasitism 1st instar, 45% in the 2nd instar and activity occurred between 72 (35% 43% in the 4th instar (Figure 9). Results parasitism) and 96 hours (33% of a second experiment were similar, parasitism) (Figure 10). However, even

321 Whiteflies and Whitefly-borne Viruses in the Tropics at 216 hours, parasitism rates still parasitized. However, whitefly remained relatively high (nearly 29%) populations remain high and cause indicating a relatively lengthy considerable plant damage and yield parasitoid activity and the need to do loss, indicating that the actual activity further evaluations at time periods of parasitoids is not sufficient to between 96 and 216 hours. reduce whitefly populations below economic injury levels that reduce cassava yields. 100

80 Discussion 60

40 In future studies, additional parasites such as Eretmocerus and Amitus species % parasitism 20 will be evaluated in similar studies to 0 those with E. hispida. Continued survey 48 72 96 216 activities to identify additional natural Time (hours) enemies also are underway. During this Figure 10. Parasitoid (Encarsia hispida) activity phase of exploration, more emphasis in relation to time of exposure to will be given to regions or fields where whitefly host Aleurotrachelus socialis on cassava in greenhouse studies. there are low whitefly populations with the hope of identifying key parasitoids, predators or entomopathogens that are The high rates of parasitism, preventing eruptions of whitefly especially using the muslin cage populations. methodology, indicate that E. hispida could be an effective parasite in a This is a collaborative project biological control program for between CIAT and the University of A. socialis. However, whitefly Florida, and is funded by the United populations, especially of A. socialis, States Agency for International have been extremely high at CIAT for Development (USAID). This the past 4 or 5 years. During this collaboration will provide training and period, E. hispida is the most improved in-country capacity for frequently observed parasite. These research, production, delivery and previously described experiments management of biological control were carried out under controlled agents. The University of Florida will conditions where the parasitoid had provide expertise and input on easy access to the whitefly prey and the parasitoid taxonomy, biology, parasitoid probably is not adversely behaviour, collecting, rearing, influenced by environmental factors. identification and data analysis. Since Under natural field conditions, 1998, University of Florida researchers parasitoid activity may not be as (Dr. Jorge Peña and R. Nguyen) have efficient as indicated by these visited CIAT and collaborated in training greenhouse studies. CIAT personnel in some of the above areas. Whitefly parasitoids that are Observations of pupal populations collected from sampling are sent to in the field at CIAT and other selected University of Florida taxonomists sites (i.e., Tolima) have shown that a (G. Evans, M. Rose and A. Hammonds) high percentage of pupae are for identification.

322 Biological Control of Whiteflies by Indigenous Natural Enemies

References Gerling, D. 1986. Natural enemies of Bemisia tabaci, biological Bellotti, A. C.; Smith, L.; Lapointe, S. L. characteristics and potential as 1999. Recent advances in cassava biological control agents: A review. pest management. Annu. Rev. Agric. Ecosyst. Environ. 17:99-110. Entomol. 44:343-70. Gold, C. S.; Altiere, M. A.; Bellotti, A. C. Castillo, J. A. 1996. Moscas blancas 1989. The effects of intercropping (Homoptera: Aleyrodidae) y sus and mixed varieties on predators enemigos naturales sobre cultivos and parasitoids of cassava de yuca (Manihot esculenta Crantz) whiteflies in Colombia: An en Colombia. M.Sc. thesis, examination of the “natural enemy Universidad del Valle, Cali, CO. hypothesis.” Bull. Entomol. Res. 173 p. 79:115-21.

CIAT (Centro Internacional de Agricultura Perring, T. M.; Cooper, A. D.; Rodriguez, Tropical). 1999. Annual Report, R. J.; Farrar, C. A.; Bellows, T. S. Project PE1. Cali, CO. 1993. Identification of a whitefly species by genomic and behavioural Costa, A. S.; Russell, L. M. 1975. Failure studies. Science 259:74-77. of Bemisia tabaci to breed on cassava plants in Brazil Quintero, C.; Cardona, C.; Ramírez, D.; (Homoptera, Aleyrodidae). Cienc. Jiménez, N. 1998. Primer registro Cult. 27:388-90. del biotipo B de Bemisia tabaci (Homoptera: Aleyrodidae) en Evans, G. A.; Castillo, J. A. 1998. Colombia. Rev. Colomb. Entomol. Parasites of Aleurotrachelus socialis 24(1-2):23-28. (Homoptera: Aleyrodidae) from Colombia including descriptions of Vargas, O.; Bellotti, A. C. 1981. Pérdidas two new species (Hymenoptera: en rendimiento causadas por Aphelinidae: Platygasteridae). Fla. moscas blancas en el cultivo de la Entomol. 81(2):171-178. yuca. Rev. Colomb. Entomol. 71(1-2):13-20. França, F. M.; Villas-Bôas, G. L.; Castelo Branco, M. 1996. Ocorrência de Bemisia argentifolii Bellows and Perring (Homoptera: Aleyrodidae) no Distrito Federal. Anais Soc. Entomol. Bras. 25:369-372.

323 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 5.3 Progress of Cassava Mosaic Disease in Ugandan Cassava Varieties and in Varietal Mixtures

William Sserubombwe*, Michael Thresh**, James Legg*** and William Otim-Nape*

A feature of the production of cassava Integrated Pest Management (TWF-IPM) (Manihot esculenta Crantz) in Uganda, Project has considered the role of as in many other countries of sub- varietal diversity in relation to CMD Saharan Africa, is that many different under epidemic conditions in Uganda. varieties are grown and it is usual to It is proposed to consider further find several varieties within individual aspects of varietal diversity and the plantings. Moreover, cassava is usually implications of intercropping in any grown together with one or more additional phase of the project. intercrops, including legumes and cereals. The great diversity in the varieties grown and in the cropping Response of Local systems adopted can be expected to Varieties Aladu and Bao influence the incidence and severity of to Cassava Mosaic Disease cassava pests and diseases. However, this possibility has received only Ugandan farmers selected the varieties limited attention in relation to cassava Aladu and Bao and originally grew mosaic disease (CMD), despite them mainly in Apac District, where indications that the current pandemic they appeared to have at least some in Uganda and in adjacent parts of degree of resistance to CMD. For this north-west Tanzania and western reason, and in the absence of more Kenya is closely associated with the resistant varieties, they were varieties adopted and is least damaging introduced to Kumi, Soroti and other in areas where many varieties are districts of Uganda in the early 1990s grown. in attempts to rehabilitate cassava production following the devastating The cassava component of the epidemic of CMD. Natural Resources Institute (NRI) contribution to the Tropical Whitefly This approach was partially successful and farmers have retained both Bao and Aladu in some areas, * Namulonge Agricultural and Animal although they are less resistant than Production Research Institute (NAARI), some of the other improved varieties Kampala, Uganda. ** Natural Resources Institute (NRI), Greenwich now available. Bao became heavily University, Chatham Maritime, Kent, UK. infected under the epidemic conditions *** International Institute of Tropical encountered originally. However, as the Agriculture-Eastern and Southern Africa epidemic abated and the symptoms Regional Center (IITA-ESARC), Kampala, Uganda, and NRI, University of Greenwich, became less severe, some farmers have Chatham Maritime, Kent, UK. attempted to return to Bao because of

324 Progress of Cassava Mosaic Disease in Single and Mixed Stands its attractive eating qualities and other earlier. Moreover, sufficient desirable attributes. symptomless plants of Aladu remained to provide cuttings for a further Replicated plots of Bao and Aladu planting. This suggests that Aladu were established in May 1998 and could be sustained even under recorded monthly to assess the epidemic conditions, provided that incidence and severity of CMD. farmers select healthy cuttings at each Comparable plots of the highly cycle of propagation. Additional studies resistant SS4 also were established as of this type are justified on the a standard. All plants of Bao and Aladu sustainability of a wider range of local were harvested individually in June varieties and under less extreme 1999 and the data for individual plants conditions of inoculum pressure than (each of which had a full set of four those encountered at Namulonge neighbours) were used for analysis. between 1998 and 1999.

There was rapid spread of CMD to Bao and all plants were affected within 120 6 months after planting (MAP) A (Figure 1). CMD spread less rapidly in 100 Aladu and 15% of the plants were unaffected at the final observation 80 12 MAP. On average, symptoms 60 developed in Aladu 5.1 MAP compared with 3.7 MAP for Bao. Few plants of 40 SS4 were affected and the symptoms Incidence of CMD (%) 20 were less conspicuous than those of Bao and Aladu. Yields of tuberous 0 roots were related to the stage of growth when symptoms were first 600 B expressed and Bao was more severely 500 affected than Aladu (Figure 2). Little yield was obtained from plants of either 400 variety that were grown from infected 300 cuttings. Plants infected by whiteflies were affected less severely, especially 200 those that developed symptoms at a 100 late stage of growth. Incidence of CMD (MIU) 0 The incidence and yield data were 0 1 2 3 4 5 6 7 8 9 10 11 12 used to calculate the losses caused by Months after planting CMD in relation to fully healthy stands Aladu SS4 Bao and assuming that all plants were Figure 1. Mean monthly incidence of cassava grown from healthy cuttings. The mosaic disease (CMD) in local overall loss was 75% for Bao compared cassava varieties SS4, Aladu and Bao at Namulonge, 1998-99 with 31% for Aladu. This confirms the experiment. Incidence data (A) as vulnerability of Bao and the relative percentages and (B) transformed to tolerance of Aladu, as farmers reported multiple infection units (MIU).

325 Whiteflies and Whitefly-borne Viruses in the Tropics

selecting from the local varieties already 5 A available within the country. Much Aladu Bao 4 information has been obtained on the performance of the resistant material 3 that has been released to farmers through official rehabilitation schemes (kg) 2 but little information is available on the features of the local varieties that have 1 been adopted widely following the epidemic. Accordingly, a study has been Tuberous root yield per plant 0 made of the 18 local varieties listed in Table 1. They were collected from 100 B 90 different parts of Uganda and selected Aladu Bao 80 for their apparent ability to withstand 70 the most damaging effects of CMD. 60 50 Wherever possible, cuttings were 40 collected from symptomless plants of

(% of healthy) 30 each variety selected for study and from 20 plants expressing mild and severe 10 symptoms. The cuttings were planted at Tuberous root yield per plant 0 Namulonge in May 1998, together with CE I LH CE I LH the varieties Ebwanateraka, Bao, Stage of virus infection Migyera and SS4, which were included Figure 2. Relationship between time of first as standards. All plants were assessed symptom expression of cassava mosaic disease and tuberous root monthly to record the incidence and yield per plant in local cassava severity of CMD. Vegetative growth and varieties Aladu (left) and Bao (right). populations of the whitefly vector Data presented as (A) mean yield per plant and (B) yields expressed as Bemisia tabaci (Gennadius) were also percentage of healthy controls. Data assessed and the plants were rated for for 10 plants each having a full set overall vigour 12 MAP. of four immediate neighbours. C = infected as cutting, E = early, I = intermediate, L = late infected Populations of adult whitefly were and H = healthy (symptomless) unusually high throughout the period of plants. (The healthy Bao were the experiment and there was rapid nominal controls sampled in an adjacent plot in the absence of spread of CMD both within the trial and uninfected plants.) in adjacent experiments. All the initially healthy plants of the local varieties became infected but there was relatively Ability of Ugandan little spread to the resistant standards. Varieties to Withstand There was wide variation in symptom Effects of Cassava Mosaic severity depending on the variety and Disease the initial health status of the planting material. Symptoms were generally Experience over the last decade has highly conspicuous except in the local shown that farmers in Uganda have varieties Buhimba, Njule and Tongolo overcome the effects of the CMD and in the resistant standards SS4 and epidemic by adopting resistant varieties Migyera. Nevertheless, many of the local introduced from Nigeria or those varieties grew vigorously, especially selected by the National Agricultural when grown from cuttings collected from Research Organisation (NARO) and by symptomless or mildly affected plants.

326 Progress of Cassava Mosaic Disease in Single and Mixed Stands Severe d Mild Vigour index severe (category 4-5) nt (MR); 41%-60%, 998-99 experiment. Data are for cted. Severe Healthy c Initial CMD status 4.0 4.8 4.8 3.2 1.4 2.0 b S 3.4 3.6 4.2 2.0 2.3 2.3 Category a Incidence (%) MIU Healthy Mild (District) MubendeMukono 85Mukono 68Mukono 190 114 65 65Mpigi HSMpigi 105 104 S 72 3.2 89 S S 2.7Mukono 129Mukono 223 4.3 4.6 4.5 68 3.7 - 48 - HS 4.9 112 5.0 4.1 64 4.2 4.0 4.8 5.0 - 4.9 S I 2.2 4.9 5.0 - 2.8 2.7 3.7 4.5 - 5.0 3.4 1.9 2.6 1.1 4.8 1.8 1.0 4.4 1.1 1.7 4.9 1.2 1.0 4.8 2.7 1.0 1.0 2.7 2.0 2.6 1.3 2.4 plants that were raised from cuttings collected symptom-less (healthy) and those with mild (category 2-3) symptoms. intermediate (I); 61%-80%, susceptible (S); 80%-100%, highly (HS). Variety Origin BuhimbaKatisa-Bakonjo Kayumba Mpologoma HoimaNjule-slUnknown 2 Kayinja MukonoBao 34Rugogoma MukonoSS4 -Luzira 41Ebwanateraka Kibaale 35Migyera SorotiAladu-aladu -Muwogo-omweru Mpigi MR 43Matooke 60 Mukono SorotiTongolo Mpigi 64Unknown 1 92Unknown 3 - 2.0 MR 0Nyaraboke Mukono 70 101 55 Masindi 13 3.2 120 - S 3.6 Masindi 78 - 79 13 S 51 3.0 150 3.8 69 S HR 4.8 72 I HR 4.0 4.3 116 4.8 S 4.9 3.1 1.0 I 4.0 2.6 3.8 2.8 2.8 4.3 - 2.6 3.7 4.4 3.6 4.6 3.0 - 2.6 5.0 4.1 1.9 - 2.0 3.2 5.0 3.7 2.5 3.9 5.0 1.3 2.2 1.3 3.8 2.1 5.0 2.1 2.1 1.9 1.2 3.4 1.2 2.0 1.1 1.9 2.6 - 1.1 1.7 1.2 1.0 Table 1. Reaction of different local cassava varieties in Uganda to mosaic disease (CMD) infection at Namulonge the 1 a. Incidence data 3 months after planting as % and transformed to multiple infection units (MIU). b. Variety categorization based on incidence 3 months after planting: 0%-20%, highly resistant (HR); 21%-40%, moderately resista c. Symptoms rated on a 1 to 5 scale of increasing severity in relation status source plants from which cuttings were colle d. Vigour was rated on a scale of 1 (least vigorous) to 5 (most vigorous), 12 months after planting.

327 Whiteflies and Whitefly-borne Viruses in the Tropics

Samples were collected from resistant and partially resistant varieties representative plants of all varieties for was similar in the sole plots and in the polymerase chain reaction (PCR) and mixture. In contrast, there was less enzyme-linked immunosorbent assay spread to the susceptible Ebwanateraka (ELISA) analysis to determine the in the mixture than when it was grown cassava mosaic virus(es) present and alone. The effect was significant and is their concentration and association likely to have been even greater under with the symptoms expressed. lower levels of inoculum pressure than Preliminary results showed that the those experienced in the two epidemic-associated virus (now experiments. This suggests that designated East African cassava mosaic resistant varieties may be used to virus-Uganda [EACMV-Ug]) and African provide at least some degree of cassava mosaic virus (ACMV) occurred protection to susceptible ones. Indeed, alone or as a mixture. The few the effect could be of considerable symptomless plants that reacted practical importance because farmers positively contained mixed infections may be reluctant to discard susceptible and so provided evidence of virus varieties that are valued because of latency. quality or other favourable attributes.

These initial studies confirm the A new varietal mixtures experiment ability of some local varieties to was planted at Namulonge in September partially withstand infection. They also in 1998. There were four treatments indicate the scope for further studies each replicated four times in a complete on the behaviour of local varieties and randomized block design: on their sustainability, especially under the less extreme conditions of (1) Variety Bao (susceptible): alone. inoculum pressure encountered in (2) Variety SS4 (resistant): alone. post-epidemic situations. (3) Bao and SS4 planted alternately in 50:50 mixture. (4) SS4 (50%) grown as an outside Spread of Cassava Mosaic barrier around Bao (50%). Disease and Whitefly Vector Populations on Monthly assessments were made of Varieties Grown Singly adult whitefly populations, CMD and as a Mixture incidence and CMD severity. The assessment of whiteflies was done so Detailed analyses have been done of that any edge or barrier effects would be field data collected during an earlier detected. M.Sc. project funded by the Rockefeller Foundation. This project assessed the The health status of the Bao spread of CMD and whitefly vector planting material was unsatisfactory populations in four contrasting and there was substantial infection varieties grown singly and as a mixture arising from the cuttings used. This under epidemic conditions at reflects the difficulty now being Namulonge in 1995-96 and 1996-97. experienced in Uganda in obtaining There was much spread of CMD to the healthy material of susceptible varieties. susceptible variety Ebwanateraka, little Nevertheless, preliminary analyses show to the resistant SS4 and intermediate big differences between SS4 and Bao in levels in TMS 30337 (Nase 2) and TMS the incidence and severity of CMD 30572 (Nase 3, also known as Migyera) (Figure 4). There was also evidence that (Figure 3). The amount of spread to the the incidence of CMD in Bao was less in

328 Progress of Cassava Mosaic Disease in Single and Mixed Stands the mixture than in the pure stand. erect variety that outgrows SS4, which is However, there was no evidence of a a relatively low spreading type. This barrier or edge effect using SS4 and the suggests that varieties of similar habit comparison may have been vitiated by and conformation should be selected for the extent of cutting infection with Bao. any further evaluation of mixtures. A further difficulty is that Bao is a tall

100 Migyera Migyera 80

100 Nase 2 Nase 2 80

Incidence of CMD (%) 0

100 Ebwanateraka Ebwanateraka 80

800 Ebwanateraka Ebwanateraka

600

400

200

Incidence of CMD (MIU) 0 4(F) 6(A) 4(S) 3(A) 1(N) 5(O) 6(N) 2(D) 7(D) 0(O) 7(M) 5(M) 0(M) 2(Jy) 8(Ja) 3(Ja) 8(Ju) 1(Ju) MAP MAP

Grown alone In varietal mixture

Figure 3. Mean monthly incidence of cassava mosaic disease (CMD, %) in successive months after planting in varieties Migyera, Nase 2 and Ebwanateraka when grown alone and as a mixture at Namulonge in (left) the 1995-96 and (right) the 1996-97 experiments. Incidence data for Ebwanateraka were transformed to multiple infection units (MIU) in the bottom two figures.

329 Whiteflies and Whitefly-borne Viruses in the Tropics

100

40 Incidence of CMD (%)

400

350

300

250

200

150 Incidence of CMD (MIU)

100

0 5(F) 5(F) 7(A) 7(A) 3(D) 3(D) 1(O) 1(O) 9(Ju) 9(Ju) MAP MAP Bao alone Bao mixture SS4 alone SS4 mixture Bao barrier SS4 barrier

Figure 4. Mean monthly incidence of cassava mosaic disease (CMD, %) in successive months after planting (MAP) in varieties SS4 and Bao when grown alone, as a mixture and with SS4 as a barrier around Bao at Namulonge in the 1998-99 experiment. Incidence expressed as % (top) and as transformed to multiple infection units (MIU) (bottom).

330 Progress of Cassava Mosaic Disease in Single and Mixed Stands

The experiment was harvested Acknowledgments 12 MAP to consider the yields of healthy and infected plants of each This study was carried out in variety in relation to the type and collaboration with the NARO Cassava healthy status of their immediate Programme, Uganda. The study will neighbours. It is particularly important continue and results will be analysed to assess the ability of SS4 to and written up with short-term funds compensate for the impaired growth of provided by the Crop Protection Bao in the mixture of the two varieties. Programme of the UK Department for International Development (DFID).

Conclusions Reference The results obtained are consistent with previous observations on the Sserubombwe, W. S.; Thresh, J. M.; behaviour of local varieties and on the Otim-Nape, G. W.; Osiru, D. O. S. 2001. Progress of cassava mosaic scope for adopting somewhat tolerant virus disease and whitefly vector varieties and varietal mixtures as a populations in single and mixed means of alleviating the effects of CMD stands of four cassava varieties (Sserubombwe et al., 2001). Further grown under epidemic conditions in studies on the behaviour and Uganda. Ann. Appl. Biol. 138: sustainability of local varieties, 161-170. especially under the less extreme conditions of inoculum pressure encountered in post-epidemic situations, need to be made.

331 Whiteflies and Whitefly-borne Viruses in the Tropics

CHAPTER 5.4 Management of the Cassava Mosaic Disease Pandemic in East Africa

James Legg*, James Whyte**, Regina Kapinga*** and James Teriψ

Introduction expansion, based on epidemiological data (Legg et al., 1999b). During the 1990s, a pandemic of an unusually severe form of cassava In view of the acute effects of the mosaic disease (CMD) expanded to severe CMD associated with the cover a large part of East Africa, pandemic, the co-ordination team of including virtually the whole of Uganda the TWF-IPM Project considered that and parts of western Kenya, southern there was an immediate need to Sudan, north-western Tanzania and identify sources of funding to support eastern Democratic Republic of Congo CMD control activities in recently (DRC) (Otim-Nape et al., 1997; affected/threatened areas. Dialogue ASARECA, 1998; Legg et al., 1999b). was initiated with staff of the United This has been associated with the States Agency for International occurrence of a novel and highly Development (USAID) by the Project virulent cassava mosaic begomovirus Co-ordinator and subsequently (Deng et al., 1997; Harrison et al., followed up in Uganda by the 1997; Zhou et al., 1997). Surveys to co-ordinator of the Africa-based assess the prevalence and severity of Sub-Project 4. A concept note was CMD were conducted in Uganda, Kenya submitted to the Office for Foreign and Tanzania as part of the diagnostic Disaster Assistance (OFDA) of USAID phase of the Tropical Whitefly in April 1998 and funding for a 1-year Integrated Pest Management (TWF-IPM) project was approved the following Project (Chapters 1.6, 1.7 and 1.8, this month. The project, entitled volume). A principal outcome of these “Emergency programme to combat the surveys was the identification of cassava mosaic disease pandemic in regions that were either currently East Africa” began in October 1998 and affected or threatened by pandemic this first phase was completed in September 1999.

* International Institute of Tropical Agriculture-Eastern and Southern Africa Background to Control of Regional Center (IITA-ESARC), Kampala, the CMD Pandemic Uganda, and Natural Resources Institute (NRI), University of Greenwich, Chatham Maritime, Kent, UK. After a decade of major losses to ** IITA, Gabarone, Botswana. cassava production in Uganda due to *** Centro Internacional de la Papa (CIP), the CMD pandemic (Otim-Nape et al., Kampala, Uganda. ψ Tanzania Coffee Research Institute, Moshi, 1997), substantial experience has been Tanzania. gained on control strategies. CMD-

332 Management of the Cassava Mosaic Disease Pandemic in E. Africa resistant cassava germplasm Pandemic control in western introduced from the International Kenya and open quarantine Institute of Tropical Agriculture (IITA) The expansion of the pandemic into in the early 1980s was evaluated in western Kenya was first noted in 1995 multi-locational and on-farm trials (Gibson, 1996; Legg et al., 1999b) during the early 1990s and, in 1993, although it was not until 2 years later three cultivars (TMS 60142, as the increasing seriousness of the TMS 30337 and TMS 30572) were problem was recognized that initiatives officially released under the names began to control it. The key constraint Nase 1, Nase 2 and Nase 3 (or Migyera). to CMD management in western Kenya In addition to being resistant to was the virtual absence of any CMD infection, they were tolerant also of the resistant germplasm in the country. In effects of cassava mosaic order to address this problem, begomoviruses (CMBs) once infected scientists of the East African Root (Thresh et al., 1994). Multiplication Crops Research Network (EARRNET) efforts were initiated during the mid- worked with plant quarantine officials 1990s (Otim-Nape et al., 1994; GCF, of both Uganda and Kenya to develop 1997) and areas initially targeted guidelines for the establishment of an included some of the first areas to be “open quarantine” facility. This was affected by the pandemic in northern sited at Busia, on the Kenyan side of and eastern Uganda (Otim-Nape et al., the Kenya/Uganda border, and allowed 1994). the introduction of stem cuttings of two CMD resistant cultivars, SS4 and In more recent years, as the TMS 30572 (Nase 3), in addition to a pandemic has expanded southwards, wide range of selected clones from the the emphasis of control efforts has EARRNET germplasm development shifted to central and southern programme, based at Serere, Uganda. districts. At the end of 1997, a major It was also at this time that the project funded by the PL 480 pandemic control programme in programme of USAID and entitled western Kenya received its first major “Dissemination and utilization of offer of financial support, from the mosaic resistant cassava in Uganda” Gatsby Charitable Foundation (GCF) in was initiated in the central-southern the UK. EARRNET co-ordinated the target districts of Masindi, Luwero, development of a project, funded by Mukono, Kamuli and Iganga. Principal GCF, which provided for the CMD management themes within this introduction, evaluation and programme were monitoring and multiplication of CMD-resistant diagnostics, plant health management cultivars in western Kenya and was within the multiplication scheme and initially targeted to run for 3 years. participatory evaluation of new CMD-resistant germplasm. The project Regional pandemic control and targeted districts that had been the OFDA CMD Project affected by the CMD pandemic for more than 1 year and therefore already Whilst USAID and Gatsby-funded were experiencing significant shortages initiatives were providing substantial of planting material. This project did support for the multiplication and not address districts affected more dissemination of CMD resistant recently, in 1996 and 1997, as cultivars in Uganda and Kenya, it was identified by the TWF-IPM Project clear that the increasing regionalization diagnostic surveys. of the CMD pandemic needed to be addressed by taking a more pro-active

333 Whiteflies and Whitefly-borne Viruses in the Tropics and co-ordinated approach. It was stakeholders with roles in envisaged that such an approach enhancing cassava production in would involve establishing links target areas. between control efforts in each of the (5) Farmer training: develop producer East African countries affected or skills in identification and about to be affected by the pandemic, management of cassava pests and and targeting monitoring and control diseases with special focus on initiatives towards “threatened” zones CMD, in addition to basic in addition to areas already affected. production and multiplication This was the rationale behind the skills. development of the OFDA-funded project, which began in October 1998. Implementation The purpose of the OFDA CMD Highlights—The OFDA Project was to “boost production of CMD Project cassava in Uganda, Kenya and Tanzania and enhance both short and Significant progress was made in the longer term food security, through the attainment of project targets in each of implementation of an emergency the main themes and the project has programme to multiply and been successful in fostering a more disseminate mosaic resistant cassava”. comprehensive regional approach to Principal partners included IITA- tackling the CMD problem. Eastern and Southern Africa Regional Center (ESARC), the two regional root Monitoring and diagnostics crops networks of EARRNET and the An important feature of the project has Southern African Root Crops Research been the use of regular monitoring and Network (SARRNET) and national root diagnostic surveys to assess the status crops programmes in Kenya, Tanzania of CMD in threatened zones and and Uganda. Principal project activities develop forecasts for the likely pattern were organized under five themes: of development of the pandemic. Surveys focused on the Kagera region in (1) Monitoring and diagnostics: north-west Tanzania and Western and conduct focused surveys in south- Nyanza Provinces in western Kenya. western Uganda, western Kenya During the first quarter of the project, and north-western Tanzania to the first report was made of severe CMD provide detailed distribution maps in Kagera and subsequent virus of CMGs in the project target areas diagnoses using specific polymerase and baseline data for subsequent chain reaction (PCR) primers confirmed impact analysis. the association of severe CMD in this (2) Multiplication: multiply and region with the presence in diseased distribute in collaboration with plants of mixed infections of African project partners, elite CMD- cassava mosaic virus (ACMV) and the resistant materials. Uganda variant of East African cassava (3) Germplasm diversification: mosaic virus (EACMV-Ug) (Legg and increase the range of cassava Okao-Okuja, 1999). The occurrence of materials available to farmers in mixed ACMV/EACMV-Ug infections is areas targeted by the project, characteristic of the “front” of the CMD thereby reducing future risk of pandemic (Harrison et al., 1997) and production collapse. the associated severe symptoms are the (4) Stakeholder linkages: identify and main reason for its acute impact on strengthen links between key cassava cultivation in affected areas.

334 Management of the Cassava Mosaic Disease Pandemic in E. Africa

In western Kenya, the distribution and used to plant 15 ha in each of of CMGs appears to be complex and Rakai and Masaka Districts. ACMV, EACMV and EACMV-Ug have been reported (Ogbe et al., 1996; Legg In western Kenya, by contrast, less and Okao-Okuja, 1999). Monitoring than 4 ha of CMD-resistant cultivars surveys in late 1998 demonstrated the were present at the outset of the apparent slowing of the spread of the project, at the open quarantine site. pandemic. This was possibly as a result The OFDA CMD Project joined GCF and of the presence of major natural the Rockefeller Foundation in barriers between north and south supporting the multiplication of this Nyanza Province (Kisumu area), which material and by June 1999 the area of include the Winam Gulf of Lake CMD-resistant material had increased Victoria and the Kano Plains, an to more than 35 ha. expanse of flood plain in which cassava cultivation is virtually absent (Legg et In north-western Tanzania, the last al., 1999a). Although high incidences of of the three countries to be affected by severe CMD have yet to be reported the pandemic, only a small number of from south Nyanza, the pandemic- plants of CMD-resistant varieties were associated EACMV-Ug has been present by the end of 1998 and detected from Migori, near the therefore more vigorous action had to Tanzania border. be taken to obtain resistant material. As a consequence, whilst lobbying the Data obtained from monitoring and Tanzania Ministry of Agriculture and diagnostic surveys have been used to Co-operatives for permission to develop regional CMD maps and their establish open quarantine in Kagera, presentation has been an important tissue culture plantlets were requested tool in raising awareness amongst from IITA, Ibadan, Nigeria, and an agricultural workers based in intensive programme of true irrigated threatened zones. It is considered also rapid multiplication (IITA, 1998) of that presentation of these graphics has available CMD-resistant cultivars was been a key component in persuading expanded at Ukiriguru Agricultural both national agricultural institutions Research Institute (ARI), Mwanza. By and donor agencies to take seriously June 1999, cassava planting material the threat posed by the pandemic to equivalent to almost half a million regional cassava production and, with cuttings had been multiplied, more it, food security. than 10,000 tissue culture plantlets of CMD-resistant cultivars had been Multiplication imported from IITA, Nigeria and proposals for the establishment of open The multiplication and dissemination quarantine had been approved. of CMD-resistant cultivars has been the main control technique used in tackling the CMD pandemic in Uganda Germplasm diversification and it has fulfilled the same role in the Broadening the genetic base of cassava OFDA CMD Project in the wider region. is seen as a key aspect of a balanced Approaches to multiplication differed in approach to the management of each of the three participating cassava pests and diseases. This is countries, however, since the particularly important with respect to availability of planting material differed the CMD pandemic, since a heavy substantially in each. In Uganda, emphasis on the development of stems of cultivar SS4 were obtained germplasm with high levels of from the USAID-funded PL 480 Project resistance to CMD has resulted in less

335 Whiteflies and Whitefly-borne Viruses in the Tropics attention being given to other suffering substantial yield losses, before important pest and disease finally abandoning the crop. Raising constraints. A practical result of this is awareness and the provision of training that the two varieties most widely in CMD management are key multiplied for CMD control in East requirements if this type of scenario is Africa, SS4 and TMS 30572, are to be avoided. Training for both relatively susceptible to cassava green extension workers and farmers mite damage. In more recent therefore has been an important part of germplasm development work, greater the OFDA CMD Project. Various fora emphasis has been placed on multiple have been used for training exercises pest and disease resistance and including: germplasm with these characteristics is now becoming available both through (1) Stakeholder workshops: held at the national breeding programmes and outset of the project in each of the through the EARRNET regional three countries. Information germplasm development programme. provided on the cause, spread, impact and strategy for the In Uganda, two demonstration sites management of the CMD pandemic. have been established in Rakai and (2) In-situ training: farmers and Masaka and include nine cultivars that extension workers have been are currently being evaluated in trained in cassava pest national on-farm trials. In Kenya, more management and multiplication than 500 clones were introduced to the techniques at multiplication sites. open quarantine site from the (3) Individual training: researchers EARRNET programme based at Serere from participating countries have and the best 15 of these were been trained in virus diagnostic multiplied for “fast-track” multi- techniques and have been locational testing in 2000. In Tanzania, sponsored to participate in regional more than 20 elite cultivars from the CMD scientific meetings. multiplication programme are under (4) Community training: village leaders evaluation in areas of the Kagera region have been trained in north-western most severely affected by the Tanzania to facilitate the transfer of pandemic, and new cultivars with novel information down to grassroots West African landrace-derived sources level. of CMD resistance formed part of the (5) Formal course training: the project recently imported tissue culture has helped sponsor courses on seed consignment from IITA, Nigeria. Each multiplication in Tanzania and on of these initiatives should help ensure cassava production, protection and that a diversity of cultivars is ultimately utilization in Kenya, principally made available to farmers that should targeted at local agricultural minimize losses likely to result from workers. future pest or disease outbreaks. In addition to the provision of Training training through courses or Changes resulting from the CMD demonstration, use has been made of pandemic are dramatic, and experience the media, particularly in Tanzania, obtained from Uganda shows that where local radio communications have farmers are typically unaware of the provided information on the CMD likely impact at the outset and pandemic to a much greater number of frequently continue with cassava households than could have been cultivation for a number of years, achieved using other methods.

336 Management of the Cassava Mosaic Disease Pandemic in E. Africa

Stakeholder linkages other countries threatened by the The key to the success the project has pandemic in the near future. Given the likely continued expansion of the achieved to date is largely attributable to the active participation of many pandemic into other neighbouring cassava stakeholders in target areas. countries, similar such vigorous responses will be essential if major These include non-governmental organizations, government district losses are not to continue across the agriculture offices, farmer training cassava-producing areas of East and Central Africa. centres and projects funded by multi- lateral donors. The inclusion of a broad range of partners has the twin advantages of fostering ownership of Acknowledgements the project at local level and enhancing The OFDA CMD Project was funded local co-ordination. The development of through a grant from the Office for a local steering committee for western Foreign Disaster Assistance of the Kenya, which was charged with the United States Agency for International responsibility for co-ordinating cassava Development. development and multiplication activities in that region, was an important development in which the OFDA Project participated (ASARECA, References 1999). ASARECA (Association for the Strengthening of Agricultural Research in Eastern and Central Conclusions Africa). 1998. Cassava mosaic pandemic threatens food security. AgriForum 3:1 and 8. In a relatively short period of time the OFDA CMD Project, affiliated to the ASARECA (Association for the TWF-IPM Project, has made significant Strengthening of Agricultural achievements in both establishing a Research in Eastern and Central regional collaborative mechanism for Africa). 1999. Fighting cassava the monitoring and management of the mosaic pandemic: Networking CMD pandemic and actually critical. AgriForum 7:1 and 12. implementing control measures. A key Deng, D.; Otim-Nape, G. W.; Sangare, A.; advantage of the pro-active approach Ogwal, S.; Beachy, R. N.; Fauquet, the project is taking, with its emphasis C. M. 1997. Presence of a new virus on targeting threatened zones, is that closely associated with cassava lead times for the implementation of mosaic outbreak in Uganda. Afr. J. control measures may be significantly Root Tuber Crops 2:23-28. reduced compared with previous GCF (Gatsby Charitable Foundation). experience in Uganda. Given the 1997. Mastering mosaic: The fight magnitude of losses associated with the for cassava production in Uganda. CMD pandemic (Legg et al., 1999b; Gatsby Occasional Paper, London, Otim-Nape et al., 1997; Thresh et al., GB. 24 p. 1997), this is an important development. The multi-faceted Gibson, R. W. 1996. The report of a approach being pursued by the survey monitoring the spread of the epidemic of African cassava mosaic Tanzania Root and Tuber Crops virus from Uganda into western Programme with its partners, and Kenya. Internal report, Natural supported by the Project, also is Resources Institute (NRI), Chatham, considered to be a useful model for GB. 17 p.

337 Whiteflies and Whitefly-borne Viruses in the Tropics

Harrison, B. D; Zhou, X.; Otim-Nape, G. Ogbe, F. O.; Songa, W.; Kamau, J. W. W.; Liu, Y.; Robinson, D. J. 1997. 1996. Survey of the incidence of Role of a novel type of double African cassava mosaic and East infection in the geminivirus-induced African cassava mosaic viruses in epidemic of severe cassava mosaic Kenya and Uganda using a in Uganda. Ann. Appl. Biol 131: monoclonal antibody based 437-448. diagnostic test. Roots 3(1):10-13.

IITA (International Institute of Tropical Otim-Nape, G. W.; Bua, A.; Baguma, Y. Agriculture). 1998. Rapid 1994. Accelerating the transfer of multiplication of cassava. A colour improved production technologies: pocketbook. Training and Extension Controlling African cassava mosaic publ. no. 122, Ibadan, NG. 61 p. virus disease epidemics in Uganda. Afr. Crop Sci. J. 2:479-495. Legg, J. P.; Okao-Okuja, G. 1999. Progress in the diagnosis and Otim-Nape, G. W.; Bua, A.; Thresh, J. M.; epidemiological characterisation of Baguma, Y.; Ogwal, S.; Semakula, cassava mosaic geminiviruses in G. N.; Acola, G.; Byabakama, B.; East Africa. Procs. VIIth Martin, A. 1997. Cassava mosaic International Plant Virus virus disease in Uganda: The Epidemiology Symposium, current pandemic and approaches 11-16 April 1999. Aguadulce to control. Natural Resources (Almeria), ES. Abstract: 74-75. Institute (NRI), Chatham, GB. 65 p.

Legg, J. P.; Kapinga, R.; Teri, J.; Whyte, J. Thresh, J. M.; Otim-Nape, G. W.; B. A. 1999a. The pandemic of Jennings, D. L. 1994. Exploiting cassava mosaic virus disease in resistance to African cassava mosaic East Africa: Control strategies and virus. Asp. Appl. Biol. 39:51-60. regional partnerships. Roots 6(1):10-19. Thresh, J. M.; Otim-Nape, G. W.; Legg, J. P.; Fargette, D. 1997. African Legg, J. P.; Sseruwagi, P.; Kamau, J.; cassava mosaic disease: The Ajanga, S.; Jeremiah, S. C.; Aritua, magnitude of the problem? Afr. J. V.; Otim-Nape, G. W.; Muimba- Root Tuber Crops. 2:13-19. Kankolongo, A.; Gibson, R. W.; Thresh, J. M. 1999b. The pandemic Zhou, X.; Liu, Y.; Calvert, L.; Muñoz, C.; of severe cassava mosaic disease in Otim-Nape, G. W.; Robinson, D. J.; East Africa: Current status and Harrison, B. D. 1997. Evidence that future threats. Akoroda, M. O.; Teri, DNA-A of a geminivirus associated J. M. (eds.). Procs. Scientific with severe cassava mosaic disease Workshop of the Southern African in Uganda has arisen by Root Crops Research Network interspecific recombination. J. Gen. (SARRNET), 17-19 August 1998, Virol. 78:2101-2111. Lusaka, ZM. p. 236-251.

338 Conclusions

Conclusions

Francisco Morales*

The Tropical Whitefly Integrated Pest al., 1990; Gibson et al., 1996), tomato Management (TWF-IPM) Project is one of (Polston and Anderson, 1997) and the most ambitious projects ever peppers (Capsicum spp. L.) (Morales and undertaken in the IPM area. The Anderson, 2001). In subtropical and magnitude of the project responds to the mountainous regions, the whitefly global distribution of the pest and the species Trialeurodes vaporariorum severe damage caused by whiteflies to (Westwood) attacks crops such as major food and industrial crops in potato (Solanum tuberosum L.), tomato, tropical and subtropical agricultural common bean, several cucurbits and regions of the world (Jones, 2003). The different vegetables (Boiteau and Singh, emergence of a more prolific and 1988; Cardona, 1995). Until recently, polyphagous biotype (B) of Bemisia cultivated grasses (Gramineae) were tabaci (Gennadius) has further among the few plant species spared by complicated the control of this pest on whitefly pests. Within the last decade, alfalfa (Medicago sativa [L.] subsp. devastating whitefly attacks have taken sativa), cotton (Gossypium hirsutum L.), place in important food and industrial common bean (Phaseolus vulgaris L.), crops such as rice (Oryza sativa L.), tomato (Lycopersicon esculentum Mill.), sorghum (Sorghum bicolor [L.] Moench), soybean (Glycine max [L.] Merr.), lettuce sugarcane (Saccharum officinarum L.) (Lactuca sativa L. var. capitata L.), peas and pastures (author’s observation). Not (Pisum sativum L.), broccoli (Brassica surprisingly, whiteflies were considered oleracea L. var. italica Plenck), collard by popular news media (e.g., CNN and (Brassica oleracea L. var. viridis L.), Newsweek) as the “pest of the 20th cabbage (Brassica oleracea L.) and okra century” and their importance as a (Abelmoschus esculentus [L.] Moench) major agricultural pest remains high in (Gruenhagen et al., 1993; Godfrey et al., the new millennium. 1995; Morales and Anderson, 2001). In developing countries of Latin America, Whiteflies harm plants in different sub-Saharan Africa, India and south- ways by extracting sap, covering plants east Asia, B. tabaci is an efficient vector with a sticky sugary substance of plant viruses affecting important food (honeydew) that affects the quality crops, particularly common bean (e.g., sticky cotton) or promotes the (Morales and Anderson, 2001), cassava growth of fungi (sooty mould) on plant (Manihot esculenta Crantz) (Fargette et surfaces blocking photosynthesis (Henneberry et al., 1996) and by transmitting viruses (Markham et al., * Centro Internacional de Agricultura Tropical 1994; Jones, 2003). In heavily infested (CIAT), Cali, Colombia. plants, hundreds of immature and adult

339 Whiteflies and Whitefly-borne Viruses in the Tropics whiteflies can be found feeding on a number of different cultivated plant single leaf, resulting in rapid plant species and adapt to new death from the massive loss of environments. Farmers that try to nutrients. Sooty mould can also escape B. tabaci in the tropical lowlands provoke plant death. And, finally, and mid-altitude (500-900 m) regions by whitefly-transmitted viruses are among growing their crops at higher altitudes the most damaging viral pathogens are often disappointed to find yet known, often causing total yield losses. another whitefly pest, Trialeurodes Genetic immunity to these viruses is vaporariorum, in the highlands. Unlike rare in most of the plant species B. tabaci, the main whitefly pest and attacked by begomoviruses. vector of plant viruses in the tropics, T. vaporariorum was considered a mere Although whiteflies have been nuisance in crops grown under associated with agriculture for controlled conditions (e.g., glasshouse, centuries, these insects were only screen-house conditions). Currently, recognized as pests in the 1950s (Ernst, T. vaporariorum is also a major pest of 1994), coinciding with the development field crops in the highlands and and intensive use of agricultural temperate regions of the world and the pesticides following World War II. Half a number of plant viruses transmitted by century later, most farmers are still not this species is steadily growing (Jones, aware of the taxonomic or biological 2003). differences that characterize different species and their physiological variants Pesticide abuse is a common factor (biotypes). Nor are they informed about in the case of the main whitefly pests, the role of these species as pests and/or regardless of the crops and ecosystems vectors of plant viruses. Therefore, affected. The excessive application of farmers apply insecticides whenever insecticides, often using inadequate they see whiteflies on their crops. In chemicals, alters the delicate balance rare instances, farmers who have between insect pests and their biological received some technical assistance control agents (i.e., predators, apply insecticides when the population parasitoids and entomopathogens). In of the pests has reached a pre- the absence of natural enemies, whitefly determined level, known as the “damage populations increase freely on threshold”. Whereas these “target” or susceptible crops and eventually “threshold” applications may be develop resistance to the most effective in controlling whiteflies as frequently used insecticides (Omer et pests (Chu et al., 1995; Riley and al., 1993; Dennehy and Antilla, 1996). Palumbo, 1995), they are totally Modern agricultural practices, such as ineffective in the case of whitefly- the intensive cropping of diverse plant transmitted viruses. Basically, few adult species that act either as suitable individuals of a whitefly vector can feeding or reproductive hosts, further transmit a virus long before its contribute to the exponential increase of population is noticed in the field or whitefly pests in disturbed reaches a particular density on environments (Godfrey et al., 1995). susceptible plants. Last but not least, climate change has played a major role in the increasing The introduction of the “B biotype” outbreaks of whitefly pests throughout of Bemisia tabaci in the Americas the world. As more forest and wild lands (Brown and Bird, 1995) has drastically are cleared for agricultural purposes, increased the capacity of this whitefly the climate becomes drier and warmer, species to cause damage to a larger conditions that shorten the life cycle of

340 Conclusions whitefly pests and thus increase their safe systemic insecticides at the proper populations (Chu et al., 1995). time. This strategy rapidly reduces the number of insecticide applications to a Small-scale farmers are trying to minimum, allowing the recovery of the grow more profitable crops in order to beneficial fauna over time. Once this maximize the output of their limited condition is met, other IPM strategies landholdings. In the absence of proper can be incorporated, such as cultural technical assistance, farmers face practices, use of entomopathogens, several problems: first, they are not physical barriers and legal measures, familiar with new crops, let alone with among others. A significant reduction in their phytosanitary problems. Exotic the use of pesticides also contributes to crops can and will encounter different lower production costs and higher pests in their new environments. Often, profits for farmers, which will become exotic pathogens and pests are critical factors for emerging economies introduced with imported plant in the age of free trade agreements. Last germplasm. In the case of horticultural but not least, men, women and children crops, and particularly vegetables, the in developing countries are new cultivars have been bred in inadvertently consuming vegetables temperate countries, even though the contaminated with highly toxic origin of some of these plant species is pesticides. The medium- and long-term the tropics (e.g., tomato and peppers). effects of these toxic residues are This means that new crops may not be difficult to determine, but they may well adapted to tropical conditions, range from chronic illness to life- including their lack of resistance to threatening ailments, including cancer. local pests. Although technical Reducing pesticide use in vegetable assistance to small- and medium-scale crops to safe levels should greatly farmers is not forthcoming, pesticide contribute to vegetables free of toxic salesmen manage to reach every corner pesticide residues, and improved health of the rural world. This explains the standards in rural and urban reliance of farmers on pesticides to populations in developing countries. protect their crops. The history of cassava mosaic The main problem of most IPM disease (CMD) is over a century old practices is their failure to control pests (Fauquet and Fargette, 1990) and it has before they cause economic damage. been almost half a century since IPM is often wrongly associated with whiteflies emerged as pests and vectors “organic agriculture”, in which chemical of plant viruses of economic importance. insecticides do not have a place. New viruses have been appearing Unfortunately, once an agricultural throughout the tropics and subtropical region has been significantly modified regions of the world with a high by the introduction of new crops, and frequency (Polston and Anderson, its biological equilibrium broken by the 1997). With the advent of molecular intensive use of pesticides, whitefly techniques, the list of whitefly-borne populations are usually too large to be viruses has increased exponentially in controlled by non-chemical IPM the last 3 decades. Altogether, practices, such as biological control considerable research has been agents or yellow traps. Thus, the TWF- conducted on these pests, and yet, their IPM Project has been following a impact seems to increase every year in different approach based on the developing countries. We have already rationalization of chemical control, discussed some of the factors that have using the most effective yet specific and contributed to the unexpected

341 Whiteflies and Whitefly-borne Viruses in the Tropics dissemination of whiteflies and whitefly- Africa, this situation has made possible borne viruses. Undoubtedly, the advent the occurrence of multiple infections of chemical insecticides and their and subsequent recombination of intensive use in crops, such as cotton, different CMD-inducing virus species elevated whiteflies to the category of (Pita et al., 2001). The recombinant pests. Another important whitefly-transmitted begomovirus has epidemiological factor, the diversification caused severe yield losses in Uganda of crops in developing countries, both as and some neighbouring countries, export commodities and cash crops, has necessitating the implementation of a also played a major role in the increase famine relief project. In the case of of these pests. common bean, the downsizing of the national and international programs The importance of some of the new that worked on the development of crops lies in their role as reproductive begomovirus-resistant bean germplasm hosts to whiteflies (Tsai et al., 1996). in Latin America has resulted in the These crops need not support very large abandonment of prime agricultural populations of whiteflies, but rather regions where common beans used to occupy large areas or regions. The best be produced. This happened because of example is perhaps soybean, which was the emergence of more pathogenic rediscovered in Latin America in the begomoviruses and the breakdown of 1970s as a potential export crop, the resistance available in the earlier increasing the total area planted to over generations of improved common bean 10 million hectares. Because individual cultivars (Morales, 2001). soybean plants do not support large populations of whiteflies, primarily B. The economic crisis that affects tabaci, soybean growers do not control most national agricultural research this insect in the soybean crop. When programs in developing nations has the soybean crops mature at the end of greatly affected the flow of technical the year, common beans are planted, information to farmers and, provoking the migration of whiteflies consequently, their capacity to manage from the soybean fields to the young complex and severe crop production common bean plantings. The search for constraints such as the whitefly more profitable crops has led small- and problem. Industrialized nations are medium-scale farmers to diversify and currently controlling whiteflies with a intensify their cropping systems, new generation of systemic insecticides providing an opportunity for whiteflies that are more efficient and selective to dispose of different food sources (Chao et al., 1997). They are available in throughout the year. developing countries, but their price is very high for most small-scale farmers, Crop improvement has played an who continue to use ineffective and important role in both the onset and highly toxic contact insecticides, often control of whitefly-related problems. In on a daily basis. These contact the case of cassava, the lack of CMD- insecticides have practically annihilated resistant cultivars and/or the limited the beneficial fauna and induced adoption of mosaic-resistant germplasm resistance in whitefly pests, giving rise has perpetuated the cultivation of CMD- to very large populations of whiteflies susceptible cassava clones (Morales, that can overcome most control 2001). Consequently, the causal viruses methods, including systemic find hosts that act as virus reservoirs insecticides. Fortunately, some of the and permanent sources of inoculum for new systemic insecticides are being the whitefly vector. In sub-Saharan produced as “generic” compounds in

342 Conclusions developing countries, at lower prices. yielded visible results in terms of raising These problems have forced many awareness about the nature of the farmers to abandon the cultivation of whitefly problem, particularly in relation susceptible crops in prime agricultural to the key issue of pesticide abuse, and regions. the use of appropriate methods to characterize whitefly pests and diagnose The TWF-IPM Project has gathered whitefly-borne viruses. The design, enough evidence to suggest that the validation, implementation and success whitefly problem can be managed with of IPM methods recommended to control the collaboration of farmers, by whiteflies and whitefly-transmitted providing proper and timely technical viruses largely depends upon a solid assistance. The challenge for the project knowledge base. is to find the most appropriate communication means to disseminate the IPM technology that has been found References to be effective and economically viable for small-scale farmers. To this end, the Boiteau, G.; Singh, P. 1988. Resistance to TWF-IPM Project is about to initiate its the greenhouse whitefly, Trialeurodes vaporariorum (Westwood) third and final phase to disseminate (Homoptera: Aleyrodidae), in a clone and scale out suitable IPM technology of the wild potato Solanum berthaultii using farmer participatory research and Hawkes. Ann. Entomol. Soc. Am. different communication channels to 81:428-431. reach affected farmers throughout the tropics, wherever whiteflies are a food Brown, J. K.; Bird, J. 1995. Variability production constraint. We are confident within the Bemisia tabaci species complex and its relation to new that well-informed farmers will be able epidemics caused by geminiviruses. to reduce the number of application of CEIBA 36:73-80. insecticides, lower production costs, increase profits, and reduce Cardona, C. 1995. Manejo de Trialeurodes environmental contamination and vaporariorum (Westwood) en frijol en human health hazards. Ultimately, la zona Andina: Aspectos técnicos, farmers will be able to recover the actitudes del agricultor y transferencia de tecnología. Memoria biological equilibrium of their IV Taller Latinoamericano sobre agricultural ecosystems, and thus moscas blancas y geminivirus, prevent future whitefly outbreaks. 16-18 octubre 1995, Zamorano, HN. CEIBA 36(1):53-64. The information presented in this book represents a major achievement in Chao, S. L.; Dennehy, T. J.; Casida, J. E. terms of compiling the international and 1997. Whitefly (Homoptera: Aleyrodidae) binding site for national (grey) literature and knowledge imidacloprid and related available on the extent and socio- insecticides: a putative nicotinic economic importance of whiteflies as acetylcholine receptor. J. Econ. pests and vectors of plant viruses. This Entomol. 90:879-882. information has been essential in selecting the most promising and viable Chu, C. C.; Henneberry, T.; Akey, D. H.; IPM strategies to be tested in Phase II of Naranjo, S. E.; Perkins, H. H.; Prabhaker, N.; Mackey, B. E. 1995. the TWF-IPM Project. The information Silverleaf whitefly: development of an contained in this book also reflects the action threshold for chemical control interest and hard work of many on cotton. In: Procs. Beltwide Cotton colleagues who now form one of the Conferences, 4-7 January 1995, most extensive research networks in the San Antonio, TX, USA. Volume 2, tropics. This joint effort has already p. 873-874. 343 Whiteflies and Whitefly-borne Viruses in the Tropics

Dennehy, T. J.; Antilla, L. 1996. Whitefly Jones, D. R. 2003. Plant viruses resistance to insecticides in transmitted by whiteflies. Eur. J. Arizona. In: Procs. Beltwide Cotton Plant Pathol. 109:195-219. Conferences, 9-12 January 1996 Nashville, TN, USA. Volume 1, Markham, P. G.; Bedford, I. D.; Liu, S. J.; p. 144-145. Pinner, M. S. 1994. The transmission of geminiviruses by Bemisia tabaci. Ernst, G. H. 1994. Whiteflies (Bemisia Pestic. Sci. 42:123-128. tabaci) in cotton: Possible origin of the problem and today’s chemical Morales, F. J. 2001. Conventional breeding control opportunities. Pestic. Sci. for resistance to Bemisia tabaci- 42:139-141. transmitted geminiviruses. Crop Prot. 20:825-834. Fargette, D.; Fauquet, C.; Grenier, E.; Thresh, J. M. 1990. The spread of Morales, F. J.; Anderson, P. K. 2001. The African cassava mosaic virus into emergence and dissemination of and within cassava fields. J. whitefly-transmitted geminiviruses in Phytopathol. 130: 289-302. Latin America. Arch. Virol. 146(3): 415-441. Fauquet, C.; Fargette, D. 1990. African cassava mosaic virus: Etiology Omer, A. D.; Johnson, M. W.; Tabashnik, epidemiology, and control. Plant B. E.; Ullman, D.E . 1993. Dis. 74:404-411. Association between insecticide use and greenhouse whitefly (Trialeurodes Gibson, R. W.; Legg, J. P.; Otim-Nape, vaporariorum Westwood) resistance to G.W. 1996. Unusually severe insecticides in Hawaii. Pestic. Sci. symptoms are a characteristic of 37:253-259. the current epidemic of mosaic virus disease of cassava in Uganda. Pita, J. S.; Fondong, V. N.; Sangaré, A.; Ann. Appl. Biol. 128:479-490. Otim-Nape, G. W.; Ogwal, S.; Fauquet, C. M. 2001. Recombination, Godfrey, L. D.; Goodell, P. B.; Summers, pseudorecombination and synergism C. G.; Bentley, W. J.; Perring, T. M. of geminiviruses are determinant 1995. Seasonal development of keys to the epidemic of severe silverleaf whitefly populations in cassava mosaic disease in Uganda. J. cotton and other crops in the Gen. Virol. 82:655-665. San Joaquin valley. In: Procs. Beltwide Cotton Conferences, Polston, J. E.; Anderson, P. K. 1997. The 4-7 January 1995, San Antonio, TX, emergence of whitefly-transmitted USA. Volume 2, p. 831-834. geminiviruses in tomato in the western hemisphere. Plant Dis. Gruenhagen, N. M.; Perring, T. M.; 81:1358-1369. Bezark, L. G.; Daoud, D. M.; Leigh, T. F. 1993. Silverleaf whitefly Riley, D. G.; Palumbo, P. J. 1995. present in the San Joaquin Valley. Abundance of two whitefly species Calif. Agric. 47:4-6. (Homoptera: Aleyrodidae) on Georgia soybean. J. Entomol. Sci. Henneberry, T. J.; Hendrix, D. L.; 30:527-533. Perkins, H. H.; Jech, L. F.; Burke, R. A. 1996. Bemisia argentifolii Tsai, J. H.; Kaihong, W.; Wang, K. H. 1996. (Homoptera: Aleyrodidae) honeydew Development and reproduction of sugars and relationships to sticky Bemisia argentifolii (Homoptera: cotton. Environ. Entomol. Aleyrodidae) on five host plants. 25:551-558. Environ. Entomol. 25:810-816.

344 Acronyms and Abbreviations

Acronyms and Abbreviations

Acronyms BMNH British Museum of Natural History ACIAR Australian Centre for BNN Banco Nacional de International Agricultural Nicaragua (National Bank Research of Nicaragua) AID Agency for International C&F New Zealand Crop and Development Food Research ANPP Association Nationale CATIE Centro Agronómico pour la Protection des Tropical de Investigación Plantes (National y Enseñanza (Tropical Association for Plant Agricultural Research and Protection), France Higher Education Centre), ARI Agricultural Research Costa Rica Institute CENSA Centro Nacional de ARP African Regional Sanidad Agropecuaria Program of AVRD (National Centre for Plant ARS Agricultural Research and Animal Health), Cuba Service of USDA, USA CENTA Centro Nacional de ASARECA Association for the Tecnificación Agrícola Strengthening of (National Centre of Agricultural Research in Agricultural Technification), Eastern and Central El Salvador Africa CGIAR Consultative Group on ASDI Agencia Sueca para el International Agricultural Desarrollo Internacional Research (Swedish Agency for CIAL Comité de Investigación International Agrícola Local (Local Development) Agricultural Research ATC Agriculture Trading Committee) Company, Malawi CIAS Centro de Investigaciones AVRDC Asian Vegetable Agrícolas del Sureste Research and (South-eastern Agricultural Development Center Research Centre), Mexico BBA Biologische CIAT Centro Internacional de Bundesanstalt für Land Agricultura Tropical und Fortwirtschaft (International Center for (Federal Biological Tropical Agriculture), Cali, Research Centre for Colombia Agriculture and Forestry), Germany

345 Whiteflies and Whitefly-borne Viruses in the Tropics

CIAZA Centro de DANIDA Danish International Investigaciones Development Agency Aplicadas a Zonas Áridas DARNDR Fondo Salvadoreño de (Research Centre for Arid Estudios de l’Agriculture, Zones), Dominican des Ressources Republic Naturelles et du CINVESTAV Centro de Développement Rural Investigaciones y (Salvadoran Fund for Estudios Avanzados studies in Agriculture, (Centre for Research and Natural Resources and Advanced Studies), Rural Development) Irapuato, Mexico DDR German Democratic CIP Centro Internacional de Republic la Papa (International DFID Department for Potato Center), Peru International CNIA Centro Nacional de Development, UK Investigación DICTA Departamento de Agropecuaria (National Investigación y Centre for Agricultural Capacitación Research), Nicaragua Agropecuaria CNRE Centre National de (Agricultural Research Recherche sur and Training l’Environnement Department), Honduras (National Centre for DIGESA Dirección General de Environmental Research), Salud Ambiental (State Madagascar Environmental Health CONAL Comisión Nacional del Office), El Salvador Algodón (National Cotton DRC Democratic Republic of Commission), Nicaragua Congo CONVERDS Collaborative Network EAP Escuela Agrícola for Vegetable Research Panamericana (Pan- and Development in American Agricultural Southern Africa School), Honduras CORPOICA Corporación Colombiana EARRNET East Africa Root Crops de Investigación Research Network Agropecuaria (Colombian ESARC Eastern and Southern Corporation of Africa Regional Center of Agricultural Research) IITA COSCA Collaborative Study of ESCaPP Ecologically Sustainable Cassava in Africa Cassava Plant CRDA Centre de recherche et Protection, regional de documentation project, Ghana agricoles (Centre for EXTOXNET Extension Toxicology Agricultural Research and Network Documentation), Haiti FAO Food and Agriculture CRS Chitedze Research Organization of the Station, Malawi United Nations, Italy CSIRO Commonwealth Scientific FCA Florida Collection of and Industrial Research Arthropods Organisation, Australia

346 Acronyms and Abbreviations

FIRA Fideicomisos Instituidos IDIAP Instituto de Investigación en Relación con la Agropecuaria de Panamá Agricultura (Agriculturally (Panamanian Agricultural Related Trusts), Mexico Research Institute) FOFIFA Centre National de IDRC International Development Recherche Appliquée au Research Centre, Canada Développement Rural IICA Instituto Interamericano (National Centre for de Cooperación para la Applied Research in Rural Agricultura (Inter-American Development), Madagascar Institute for Cooperation on FSCA Florida State Collection of Agriculture) Arthropods IIE Institution of FUNDESA Fundación para el International Education, Desarrollo Agropecuario UK (Foundation for IITA International Institute of Agricultural Development), Tropical Agriculture, Santo Domingo, Nigeria Dominican Republic IITA-ESARC International Institute GCF Gatsby Charitable of Tropical Agriculture- Foundation, UK Eastern and Southern GCRIO Global Change Research Regional Center, Nigeria Information Office, USA INIA Instituto Nacional de GIIT Grupo Interinstitucional Investigación Agraria e Interdisciplinario de (National Institute of Tomate (Interinstitutional Agricultural Research), and Interdisciplinary Mexico Tomato Group), Nicaragua INIAP Instituto Nacional GTZ Deutsche Gesellschaft Autónomo de für Technische Investigaciones Zusammenarbeit Agropecuarias (National (German Agency for Institute for Agricultural Technical Cooperation) Research), Ecuador HORTI Horticultural Research INIFAP Instituto Nacional de and Training Institute, Investigaciones Tanzania Forestales y Agropecuarias HPR Host Plant Resistance (National Institute for Project of MFAT Agricultural and Forestry IARCs International Agricultural Research), Sinaloa, Mexico Research Centres INISAV Instituto de ICA Instituto Colombiano Investigaciones de Agropecuario (Colombian Sanidad Vegetal (National Institute of Agriculture Institute for Research on and Livestock) Plant Health), Havana, ICIPE International Centre of Cuba Insect Physiology and INRAB Institut National de Ecology, Kenya Recherche Agronomique ICTA Instituto de Ciencia y du Bénin (Benin National Tecnología Agrícolas Institute for Agricultural (Institute of Agricultural Research) Science and Technology), Guatemala City, Guatemala 347 Whiteflies and Whitefly-borne Viruses in the Tropics

IOBC International NARES National Agricultural Organisation for Research and Extension Biological and Integrated Systems Control for Noxious NARO National Agricultural Animals and Plants Research Organisation, IRA Institut de Recherche Uganda Agronomique NORAD Norwegian Agency for (Agricultural Research Co-operation for Institute), Cameroon Development ISA Instituto Superior de NRCRI National Root Crops Agricultura (Higher Research Institute, Institute for Agriculture), Nigeria Dominican Republic NRI Natural Resources ISTRC-AB International Society for Institute, UK Tropical Root Crops- NRTCIP National Root and Tuber Africa Branch Crops Improvement JIC John Innes Centre, UK Project, Ghana KARI Kenya Agricultural OFDA Office for Foreign Research Institute Disaster Assistance of LILIANA Instituto de USAID Investigaciones de OIRSA Organismo Internacional Hortícolas “Liliana Regional de Sanidad Dimitroya” (Horticultural Agropecuaria (Regional Research Institute International Organization “Liliana Dimitroya”), Cuba for Plant Protection and LZARDI Lake Zone Agricultural Animal Health), Research and Nicaragua Development Institute, PAUP Phylogenetic Analysis Tanzania Using Parsimony MAG Ministerio de Agricultura PCCMCA Programa Cooperativo y Ganadería (Ministry of Centroamericano para el Agriculture), Nicaragua Mejoramiento de Cultivos MARNDR Ministère de l’agriculture, Alimenticios des ressources naturelles (Centroamerican et du développement rural Cooperative Programme for (Ministry of Agriculture, the Improvement of Food Natural Resources and Crops), Guatemala Rural Development), Haiti PHYLIP Phylogeny Inference MFAT Ministry of Foreign Package Affairs and Trade, New PNUMA Programa de las Naciones Zealand Unidas para el Medio MIP Manejo Integrado de Ambiente (United Nations Plagas (Integrated Pest Environment Programme), Management) Geneva MSU Montana State University, PPRI Plant Protection Research USA Institute, South Africa NAARI Namulonge Agricultural PPRSD Plant Protection and and Animal Production Regulatory Services Research Institute, Directorate, Ghana Uganda

348 Acronyms and Abbreviations

PROFRIJOL Programa Cooperativo TRTCP Tanzania Root and Tuber Regional de Frijol para Crops Programme Centro América, México TWF-IPM Tropical Whitefly y el Caribe (Regional Co- Integrated Pest operative Bean Management Project Programme for Central UA University of Arizona- America, Mexico and the Tucson, USA Caribbean) UFL University of Florida- PROMIPAC Programa Regional de Gainesville, USA Manejo Integrado de UNA Universidad Nacional Plagas en América Agraria (National Central (Regional Agricultural University), Integrated Pest Nicaragua Management Program for UNDP United Nations Central America) Development Programme REDCAHOR Red Colaborativa de USAID United States Agency for Investigación y International Desarrollo de Hortalizas Development para América Central, USDA United States Panamá y República Department of Dominicana Agriculture (Collaborative Network for USNM United States National Vegetable Research and Museum Development for Central UTLA Universidad Técnica America, Panama and Latinoamericana (Latin Dominican Republic) American Technical SADC Southern Africa University) Development Community UW University of Wisconsin- SARH Secretaría de Agricultura Madison, USA y Recursos Hidráulicos (Secretariat of Agriculture Abbreviations and Hydraulic Resources), Mexico AC1 replicase SARI Selian Agricultural ACMD African cassava mosaic Research Institute, disease Tanzania ACMV African cassava mosaic SARRNET Southern African Root virus Crops Research Network A.f. Aleurothrixus floccosus SDC Swiss Agency for AFLP amplified fragment Development and length polymorphism Cooperation ANOVA analysis of variance SEA Secretaría del Estado de A.p. Aleyrodes proletella Agricultura (State AV1 capsid protein Agricultural Secretariat), B.a. Bemisia afer Dominican Republic BCaMV Bean calico mosaic virus SENASA Servicio Nacional de BDMV Bean dwarf mosaic virus Salud Animal (National BG begomovirus Service for Animal BGMV Bean golden mosaic virus Health), El Salvador BGYMV Bean golden yellow SP-IPM Systemwide Programme mosaic virus on Integrated Pest B.h. Bemisia hirta Management BMMV Bean mild mosaic virus 349 Whiteflies and Whitefly-borne Viruses in the Tropics

BR Brazil ICMV Indian cassava mosaic B.t. Bemisia tabaci virus B.t.-A Bemisia tabaci A biotype IPM integrated pest B.t.-B Bemisia tabaci B biotype management C plant infected as cutting Is Israel CAR carbamates ITS internal transcribed CLCuV Cotton leaf curl virus spacers CMD cassava mosaic disease L late infected plant CMG cassava mosaic LC lethal concentration geminivirus LSD least significant CMV cucumoviruses, difference Cucumber mosaic virus MABs monoclonal antibodies CO Colombia MAB-BS a broad spectrum COI cytochrome oxidase I monoclonal antibody CPMMV Cowpea mild mottle virus used to detect bi-partite CVYV Cucumber vein yellowing begomoviruses virus MAB-GA a monoclonal antibody DAS double antibody used to detect the sandwich original Middle American DNA deoxyribonucleic acid isolates of Bean DR Dominican Republic golden yellow mosaic DSM Dar es Salaam virus-Guatemala E early infected plant MAP months after planting EACMV East African cassava MIU multiple infection units mosaic virus Mo Mochis Eg Egypt MR moderately resistant EIL economic injury level MYMV Mungbean yellow mosaic ELISA enzyme-linked virus immunosorbent assay NARS national agricultural EM electron microscopy research systems EMIF electronic monitoring of NCM nitrocellulose membrane insect feeding nt not tested EPG electrical penetration O.c. Orchamoplatus citri graph OLCV Okra leaf curl virus ES El Salvador OPs organophosphate FFS farmer field school insecticides FL fiducial limits ORF open reading frame GA Guatemala PAGE polyacrylamide gel GIS geographic information electrophoresis systems PCR polymerase chain GV geminiviruses reaction H healthy, symptomless PepGMV Pepper golden mosaic plant virus HD Honduras PepLCV Pepper leaf curl virus HPR host plant resistant PHYVV Pepper Huasteco yellow HR highly resistant vein virus HS highly susceptible PR Puerto Rico I intermediate infected PTY1 monoclonal antibody to plant detect potyviruses I intermediate resistance PVY Potato virus Y

350 Acronyms and Abbreviations

PYR pyrethroids Tan Tanzania QTL quantitative trait locus TAS triple antibody sandwich R whitefly resistant cultivar TBR tree-bisection RAPD random amplified reconnection polymorphic DNA TEV Tobacco etch virus RF resistance factor (Potyviridae: Potyvirus) RFLPs restriction fragment TLC tomato leaf curl disease length polymorphisms TLCV Tomato leaf curl virus, RH relative humidity now ToLCV S susceptible TMV Tobacco mosaic virus S whitefly susceptible (Tobamovirus) cultivar ToLCV Tomato leaf curl virus SACMV South African cassava ToMoTV Tomato mottle Taino virus mosaic virus ToMoV Tomato mottle virus SCAR sequence characterized ToYMoV Tomato yellow mottle amplified region virus

SEA East African strain ToYMV Tomato yellow mosaic SiGMV Sida golden mosaic virus virus SLCV Squash leaf curl virus TPV Texas pepper virus S.p. Siphoninus phillyreae (= Pepper golden mosaic SPCFV Sweetpotato chlorotic virus) fleck virus T.r. Trialeurodes ricini SPCSV Sweetpotato chlorotic T.v. Trialeurodes vaporariorum stunt virus TVTV Tomato vein thickening SPFMV Sweetpotato feathery virus mottle virus TYLCV Tomato yellow leaf curl SPLV Sweetpotato latent virus virus SPMMV Sweetpotato mild mottle Ug Uganda virus WmCSV Watermelon chlorotic SPSV-SEA East African strain of stunt virus SPCSV WTVs whitefly-transmitted SPVD sweetpotato virus viruses disease Xam Xanthomonas axonopodis T whitefly tolerant cultivar pv. manihotis T.a. Tetraleurodes andropogon

351