Rice tungro disease management - IRRI books - International Rice ...

RiceTungro DiseaseManagementEdited byT.C.B. Chancellor, O. Azzam, and K.L. Heong1999IRRIINTERNATIONAL RICE RESEARCH INSTITUTE

The International Rice Research Institute (IRRI) was established in 1960 by the Ford and Rockefeller Foundationswith the help and approval of the Government of the Philippines. Today IRRI is one of 16 nonprofit internationalresearch centers supported by the Consultative Group on International Agricultural Research (CGIAR). TheCGIAR is cosponsored by the Food and Agriculture Organization of the United Nations (FAO), the InternationalRank for Reconstruction and Development (World Bank), the United Nations Development Programme(UNDP), and the United Nations Environment Programme (UNEP). Its membership comprises donor countries,international and regional organizations, and private foundations.As listed in its most recent Corporate Report, IRRI receives support, thrugh the CGIAR, from a numberof donors including UNDP, World Bank, European Union, Asian Development Bank, and RockefellerFoundation, and the international aid agencies of the following governments: Australia, Belgium, Canada,People's Republic of China, Denmark, France, Germany, India, Indonesia, Islamic Republic of Iran, Japan.Republic of Korea, The Netherlands, Norway, Peru, Philippines, Spain, Sweden, Switzerland, Thailand, UnitedKingdom, and United States.The responsibility for this publication rests with the International Rice Research Institute.The designations employed in the presentation of of the material in this publication do not imply theexpression of any opinion whatsoever on the part of IRRI concerning the legal status of any country, territory,city, or area, or of its authorities, or the delimitation of its frontier, or boundaries.Copyright International Rice Research Institute 1999Los Baños, PhilippinesMailing address: MCPO Box 3127, Makati City 1271, PhilippinesPhone: (63-2) 845-0563, 844-3351 to 53Fax: (63-2) 891-1292, 845-0606Email: IRRI@CGIAR.ORGTelex: (ITT) 40890 Rice PM; (CWI) 14519 IRILB PSCable: RICEFOUND MANILAHome page: http://www.cgiar.org/irriRiceweb: http://www.riceweb.orgRiceworld: http://www.riceworld.orgCourier address: Suite 1009, Pacific Bank Building6776 Ayala Avenue, Makati City, PhilippinesTel. (63-2) 891-1236, 891-1174, 891-1258, 891-1303Suggested citation:Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the lnternational Workshop on Tungro Disease Management, 9-11 November1998, IRRI, Los Baños, Laguna, Philippines. Makati City (Philippines): International RiceResearch Institute. 166 p.EDITORS: Katherine Lopez and Bill HardyPAGE MAKEUP AND COMPOSITION: Grant Leceta and Arleen RiveraARTWORK: Emmanuel PanisalesCOVER DESIGN: Juan LazaroCOVER PHOTOGRAPHY: Ossmat AzzamISBN 971-22-0031-0This book is an output from a project funded by the UK Department forInternational Development (DFID) for the benefit of developing countries. Theviews are not necessarily those of DFID (R6519 Crop Protection Programme).

ContentsRice tungro disease in the PhilippinesX.H. Truong, E.R. Tiongco, E.H. Batay-an, S.C. Mancao, M.J. Du,and N.A. JuguanPreliminary analysis of genetic variation of rice tungrobacilliform virus in two provinces of the PhilippinesM. Arboleda, F. Sta. Cruz, and O. AzzamPreliminary analysis of genetic variation of rice tungrospherical virus in the PhilippinesK.L.M. Umadhay, M.L.M. Yambao, and O. AzzamBreeding for rice tungro disease resistance at PhilRiceL.S. Sebastian, E. R. Tiongco, D.A. Tabanao, G. V. Maramara, S. Abdula,and E.R. TabelinBreeding for rice tungro virus resistance in IndonesiaA.A. Daradjat, I.N. Widiarta, and A. HasanuddinGenetic engineering of rice for tungro resistanceO. Azzam, A. Kloti, F. Sta. Cruz, J. Fütterer, E.L. Coloquio, l. Potrykus,and R. HullMultilocation evaluation of advanced breeding lines forresistance to rice tungro virusesR.C. Cabunagan, E.R. Angeles, S. Villareal, O. Azzam, P.S. Teng,G.S.Khush. TC. B. ChanceIlor, E.R. Tiong.co, X.H. Truong. S. Mancao.I.G.N. Astika, A. Muis, A.K. Chowdhury, V. Narasimhan,T. Ganapathy, and N. SubramanianProspects of virus-resistant varieties for controlling rice tungrodisease in BaliI. G. N. AstikaEvaluating rice germplasm for resistance to rice tungrodisease in West Bengal, IndiaA.K. ChowdhuryRice tungro disease resistance and management inTamil Nadu, IndiaN. Subramanian, T. Ganapathy, M. Surendran, and A. Azeez Basha1111723313945596771iii

Tungro screen kits for extension agents and plant breedersO. Azzam, L. Kenyon, and P.D. NathAre tungro disease counts repeatable?K.G. SchoenlySurveillance scheme for tungro forecasting in MalaysiaA.B. Othman, M.J. Azizah, and A.T. JatilFarmers’ rice tungro management practices inIndia and the PhilippinesH. Warburton, S. Villareal, and P. SubramaniamCommunity-based rice pest managementX. H. Truong, E. H. Batay-an, S. C. Mancao, G. N.A. Abrigo,A.B. Estoy, L.B. Flor, Jr., H.D. Justo, Jr., E.R. Tiongco,R.N. Casco, and S.R. ObienThe influence of varietal resistance and synchronyon tungro incidence in irrigated riceecosystems in the PhilippinesT.C. B. Chancellor, E. R. Tiongco, J. Holt, S. Villareal,and P.S. TengImproving IPM technology for rice tungro diseasein IndonesiaA. Hasanuddin, I.N. Widiarta, and YuliantoLeafhopper control by insecticides is not the solutionto the tungro problemS. VillarealThe role of vector control in rice tungro disease managementN. WidiartaGLH control for the management of rice tungro diseaseT. Ganapathy, N. Subramanian, and M. SurendranManagement of rice tungro disease by chemical controlof the green leafhopper vectorE. H. Batay-an and S.C. Mancao77818593105121129139143153159iv

ForewordThe intensification of rice production that began in many countries in Asia during the1960s resulted in clear benefits for both producers and consumers. Significantly highergrain yields led to increased incomes for farmers and the production gains met theneeds of rapidly expanding populations. The new technologies that were associatedwith this process, however created certain unintended consequences. Some pests anddiseases that were previously considered to be unimportant suddenly became majorthreats to the stability of rice production. Large-scale outbreaks of an unknown virusdisease caused massive production losses and seriously affected the livelihoods ofrice farmers. This disease, which came to be known as “tungro,” remains a seriousproblem in intensively cultivated irrigated areas in South and Southeast Asia.The traditional approach to managing rice tungro disease has been through theuse of insecticides to control the leafhopper vector and through the deployment ofresistant varieties. The adverse effects of insecticide applications on human healthand on the environment, however, are well documented and this approach has severelimitations. The development of resistant varieties, in which IRRI played a leadingrole, helped to reduce tungro incidence but in some areas the resistance was shortliveddue to shifts in the virulence of the vector. Consequently, it was recognized adecade ago that new approaches to managing tungro were needed.IRRI has long sought to embrace and promote an ecological approach to pestmanagement. Integrated pest management (IPM) uses a combination of control methodsto achieve a reduction in insect numbers or disease incidence to levels at whichcrop damage is minimized. In line with its efforts to help farmers reduce their dependencyon insecticides, IRRI has been actively researching appropriate IPM strategiesfor tungro disease management. Host-plant resistance remains the cornerstone of suchstrategies and an exciting new development is the availability of advanced lines withresistance to rice tungro viruses. The application of biotechnology to breeding fortungro resistance offers additional opportunities for using diverse mechanisms of resistance.Nevertheless, recent findings on the genetic variability of tungro viruses callfor the strategic deployment and monitoring of genes for tungro resistance.This proceedings summarizes recent research results presented by plant breeders,entomologists, virologists, and sociologists at a workshop on tungro disease managementheld in Los Baños on 9-11 November 1998. Much of this work was conductedby scientists from IRRI and the Natural Resources Institute of the Universityof Greenwich in collaboration with partners from India, Indonesia, and the Philippines.Significant progress has been made in key areas of research and studies havebeen undertaken to ensure that research outputs meet the real needs of rice farmers.Understanding how farmers view tungro, and which factors influence their diseasemanagement decisions, is crucial to the development of effective control strategiesand much useful information has been gained in this area. It was also encouraging towitness in this workshop a move from tactical control methods such as insecticidesprays to strategic approaches to pest management. Coordinated management strate-V

gies implemented by groups of farmers that take into account the whole spectrum ofpest and disease problems in a given locality are needed if substantial impact is to beachieved. These changes reflect the clear commitment of IRRI and its partners toincrease yields and protect farmers’ livelihoods through sustainable production technologies.IRRI is grateful to the Department for International Development of the UnitedKingdom for financial support, which was provided through the Crop ProtectionProgramme under its Renewable Natural Resources Knowledge Strategy.RONALD P. CANTRELLDirector GeneralInternational Rice Research Institutevi

PrefaceProviding farmers with options to make rational decisions to manage tungro has beenthe primary goal of recent research conducted by IRRI in collaboration with partnersat the Natural Resources Institute in the United Kingdom and in various researchinstitutes in South and Southeast Asia. A workshop on rice tungro disease managementwas held on 9-11 November 1998 at IRRI headquarters. The objective was toreview the progress made in these research activities conducted in the region in thelight of the array of crop management constraints faced by rice farmers. Farmers’varietal preferences and the difficulty they face in modifying cultural practices suchas planting dates are highlighted in two papers as being of particular importance indeveloping appropriate tungro management strategies. Similarly, the need to integratetungro control measures into an overall pest management strategy practiced atthe community level is emphasized.The development and field evaluation of rice varieties with resistance to ricetungro viruses has been a major focus of recent research as reported by papers describingresistance breeding programs in the Philippines and Indonesia. The potentialimpact of new virus-resistant lines in reducing tungro incidence in endemic areas isconvincingly demonstrated in papers summarizing field trials in these two countriesand in India. A cautionary note is sounded in a paper that shows the degree of variationthat exists in tungro viruses at the molecular level and suggests that serious considerationneeds to be given to how best to deploy the genes with virus resistance.Further breeding work is needed to improve the agronomic characteristics of some ofthe most promising elite lines, but one line is currently being evaluated by the PhilippineSeedboard for release as a stopgap variety for endemic areas.Genetic engineering approaches to complex diseases, such as tungro, offer anability to transfer single genes without any linkage to other traits and provide anopportunity for introducing novel genes that might increase the durability of resistanceand improve agronomic characteristics. In collaboration with IRRI. several institutionshave explored these approaches and rice varieties such as IR64, TN1, TP309,and Kinuhikari were successfully transformed with various antiviral strategies tocontrol tungro. Evaluation of 71 transgenic lines is reported here and results showthat effective protection mechanisms must be directed against highly conservedfunctions or sequences to be successful. Such findings stress the need for a betterunderstanding of the molecular biology of both viruses–rice tungro bacilliform virusand rice tungro spherical virus.The role of vector control in tungro management was a lively subject of debate atthe workshop. Opinions remained divided, as can be seen from the different perspectivespresented in the papers on this subject. It was agreed, however, that seedbedprotection was only justified in specific cases and that category 1 insecticides shouldnot be used for leafhopper control under any circumstances. It is encouraging to notethat leafhopper control through insecticides is no longer a central plank in tungromanagement strategies recommended for Indonesia and the Philippines, countrieswhere the disease remains a persistent problem.vii

The development of a simple, robust prototype screening kit for rice tungro bacilliformvirus was reported in one paper. This tool will greatly facilitate tungro resistancebreeding programs and has further applications for epidemiological studies andfor field monitoring. Optimization of the screening kit and application of the techniquefor rice tungro spherical virus diagnosis were seen by workshop participants asfuture research priorities. Another fertile area that was identified for future work wasthe mapping of virus resistance genes and the use of marker-aided selection to advanceresistance breeding programs.We hope that this book will provide a resource for researchers and practitionersin the field of rice pest management and that it will stimulate further debate on themost appropriate areas for future research. We acknowledge the contribution of allthe participants at the workshop and thank Ms. Ellen Genil for her assistance in organizingthe meeting and in typing the manuscripts.T.C.B. CHANCELLOR, O. AZZAM, and K.L. HEONGviii

Rice tungro disease in the PhilippinesX.H. Truong, E.R. Tiongco, E.H. Batay-an, S.C. Mancao, M.J.C. Du, and N.A. JuguanThe rice tungro disease scenario in the Philippines, research trends, andcurrent extension activities are briefly explained and discussed in this paper.Farmers’ perceptions and knowledge on the subject were also includedbecause farmers make decisions for sustaining farm income and protectingthe agroecosystem. Some farmer groups have used green leafhopper(GLH)-resistant varieties for the past several years through the SeedNetand Farmers' Field School, yet rice tungro disease (RTD) remains a majorconstraint to rice production. For the first time in rice technology development, rice tungro virus (RTV)-resistant advanced breeding lines were identified.A few farmers in RTD-endemic areas evaluated these lines on theirfarms. Seeds of these lines with yields of 3.5–4.5 t ha -1 are being increasedand will be deployed as stopgap materials for farmers during thewet season. Some progress had been made in understanding the genotypesof tungro viruses present during recent RTD outbreaks, and the roleof cropping practices and infective GLH or virus inoculum during the pastRTD outbreak in Mindanao. More collaborative efforts in research andextension have to be undertaken to strengthen farmers’ decision makingfor appropriate RTD management in the context of a sustainable farmingsystem.RTD outbreaksThe early outbreaks of rice tungro disease in the 1940s in major rice-growing regionsin the Philippines had reduced yield by 1.4 million t annually (Serrano 1957), longbefore the causal organism of the disease was known (Hibino et al 1978, 1991), Majoroutbreaks in 1962, 1969, 1971, 1975, 1977, 1984, 1986, and 1989 in areas associatedwith intensive cultivation of early recommended varieties were reported by municipalagricultural officers (MAO) of the Department of Agriculture (DA), and reviewedby Baria (1997). Recent outbreaks occurred sporadically during 1993-98, affectingareas from 900 to 2,700 ha yearly in Mindanao (Table 1). In addition, about700 ha were affected in Negros Occidental in 1998. Crop loss estimated by the DAprovincial agricultural office from the outbreak in Davao del Norte in 1993 alonereached – P10.6 million (US$406,494). Most of the affected locations commonly reportedwere irrigated lowlands under climatic types III and IV. Dry and wet seasonsin type III are not pronounced; it is relatively dry from November to April and wetduring the rest of the year, whereas rainfall in type IV (>150 mm monthly) is more orless evenly distributed throughout the year. In most cases, rice tungro disease (RTD)outbreaks were attributed to insufficient irrigation water, which prompted farmers topractice staggered planting. The consolidated report of the DA-MAO showed thatearly infection at the vegetative stage of IR64 and IR60, BPI Ri10, PSB Rc4, 6, 8, 12,20, Masipag, and Bugos induced severe crop losses varying from 39% to 65%.No comprehensive information on the agroecosystem during the reported RTDoutbreaks was available. The only known fact was that recent outbreaks occurredduring August-October. The average monthly rainfall during the same period recorded

Table 1. Profiles of major rice production areas in Mindanao and outbreaks of rice tungro disease(RTD) in the 1993-98 wet seasons a .Location Province Area harvested Production Average yield RTD Wet season(ha) (t) (t ha -1 ) outbreak Aug-OctEndemicCotabato basinAgusan basinValencia/Pulangi basinNorth CotabatoSultan KudaratSouth CotabatoMaguindanaoDavao del NorteAgusan del SurAgusan del NorteBukidnon42,03038,05036,14052,70028,55010,8809,46053,510132,993121,80798,54491,93193,59833,82226,920188,3153.23.22.71.73.33.12.93.5Kapatagan areaEast LakeshoreBanay-Banay areaHagonoy areaTagao valleySibuguey River basinLanao del NorteLanao del SurDavao OrientalDavao del SurSurigao del SurZamboangadel Sur17,89014,3904,32012,6606,33053,12063,27233,24316,35647,69516,205198,0793.52.33.83.82.63.71,000–8972,0002,7021,2002,00019961993199619981997a From "Strategic Rice Areas in Research and Development for Mindanao" and consolidated reports from the Departmentof Agriculture-Provincial and Municipal Agricultural Offices.1997at the Philippine Rice Research Institute (PhilRice) experiment station at Midsayap,North Cotabato, for the past 18 years ranged from 170 to 250 mm, which favors greenleafhopper movement. Initial data from light traps at this station showed that theGLH population peak coincided with this rainfall level. Monthly rainfall higher than250 mm or lower than 100 mm gradually decreased the population. Vector populationand infective GLH in relation to weather factors in the RTD outbreak areas must becontinuously monitored over the year to obtain a better disease forecast. In an analysisof historical survey data from RTD-endemic areas in Mindanao, Savary et al (1993)reported that a high GLH population coupled with a high proportion of viruliferousvectors associated with specific cropping practices increased RTD incidence. In additionto these variables, the epidemic role of the presence of genotypic variations amongrice tungro spherical virus (RTSV) and rice tungro bacilliform virus (RTBV)(Cabauatan and Koganezawa 1994, Arboleda et al 1997, Yambao et al 1997, Villegaset al 1997) in the endemic areas must be investigated. De los Reyes et al (1998) alsofound mixed infections and variations in RTBV genotypes during the 1997 tungrooutbreaks in Zamboanga del Sur and Lanao del Norte and during the 1998 outbreaksin Negros Occidental and Bukidnon. Identification of the selection factors that influencevirus variation in the agroecosystem may be the key to RTD management.Farmers' perceptions, knowledge, and control practices of RTDRTD is considered as the most important factor limiting rice production by mostfarmers in endemic areas. Survey information from 658 farmers in five RTD-endemicand RTD-nonendemic provinces (Albay, Davao del Norte, Davao del Sur, North2 Truong et al

Cotabato, and Laguna) during 1995-97 showed that most farmers had experiencedthe RTD problem and could recognize symptoms of RTD (Warburton et al 1996,1997, Truong et al, this volume). Only a few farmers (8%), however, were aware thatGLH is the vector of RTD. Most of them did not know the relationship between thedisease infection and the vector spreading the disease from an infected crop to a healthyone. Consequently, they were not aware of the risks posed by a nearby infected crop,which serves as a source of disease inoculum. They usually associated the spread ofthe disease to factors such as kind of insect on the crop, water, soil, rain, and others.Some confused RTD with nutrient deficiency symptoms because the causal organismsare not observable. Their assessment of crop losses was based on past cropfailures. Control practices focused more on preventive measures for all kinds of insectpests because farmers know more about insects than RTD. Most farmers. whethertrained or untrained on tungro management technologies in “hot spots,” intensivelyused insecticides as a tool against RTD and other pests. Cypermethrin, a pyrethroidcompound, was effective against GLH in experimental fields (Batay-an and Mancao’this volume). Its effectiveness, however, was not always observed by farmers becauseRTD infection had spread. This control strategy aimed more at producing aclean crop than making farming sustainable although some realized that insecticideuse was not always effective in controlling RTD.Meanwhile, most farmers were reluctant to remove the infected crop becausethey had already incurred costs in purchasing production inputs. In Negros Occidental,farmers preferred to broadcast salt in RTD-infected fields although this practicehas no scientific basis, while others practiced roguing infected plants. Tiongco et al(1998), however, pointed out that roguing as a tactical means of control was not effectivebecause it is usually done late, when infected plants are observed. Infectedplants without symptoms remain in the field and serve as virus sources. It is thus achallenge for research and extension workers to look for an alternative control techniqueand to help strengthen farmers’ skills and efficiency in making decisions onRTD management.On the other hand, farmers in RTD-endemic areas in North Cotabato are verykeen on selecting varieties and establishing the crop for RTD management (Truong etal, Community-based rice pest management. this volume). They have more experiencewith direct seeding (DS), and they claim that a crop established by this methodhas a reduced risk of pest infestation compared with transplanting. Trained farmersfrom the Farmers’ Field School (FFS) had adopted GLH-resistant varieties such asIR56, IR62, PSB-Rc 10, PSB-Rc 18, and PSB-Rc 34 to control RTD. Untrained farmers preferred their selections.Current trends in RTD research and extensionThe use of resistant varieties is a cost-effective component of RTD management. ThePhilRice Varietal Improvement Program and the Department of Agriculture have focusedon developing varieties with high yield and good quality and, at the same time,resistance to insect pests and diseases. Baria (1997) summarized the protocols onvarietal selection of the National Cooperative Test network and approval of recom-Rice tungro disease 3

mended varieties by the National Seed Industry Council of the Philippines. From1968 to 1993, 65 varieties recommended for all rice ecosystems were developed byIRRI, Bureau of Plant Industry, the University of the Philippines Los Baños, andPhilRice (Padolina 1995). GLH-resistant varieties were usually selected (Khush 1977,1989). Their reactions to RTD in hot spots had changed, however, because of changesin GLH virulence (Dahal et al 1990). Among the 44 varieties recommended for theirrigated lowland, seven GLH-resistant varieties (IR56, IR62, IR72, IR74, PSB Rc10, PSB Rc 18, and PSB Rc 34) were commonly planted by farmers in RTD-endemicareas. IR56 had the least RTD incidence (2-12%). Only a few farmers adopted IR56,however, because it was susceptible to bacterial leaf blight. As expected, IR64 iswidely planted because of its good eating quality and high price in the market althoughit is now susceptible to RTD.After an immunological test for RTD indexing was established in 1985 (Bajet etal 1985), considerable progress was made in identifying rice accessions with resistanceto RTSV (Hibino et al 1987, 1991). Since then, rice tungro virus resistance inUtri Merah (two recessive genes) and Utri Rajapan (one recessive gene), and its eightbreeding lines resistant to RTSV has been identified (Sebastian et al, this volume).Promising breeding lines were also derived from wild rice species Oryza rufipogonand O. brachyantha (Alfonso et al 1996). Likewise. three advanced breeding lines—IR69705-1-1-3-2-1, IR69726-116-1-3, and IR71031-4-5-5-1—among several otherswith the donor parents of Utri Merah, ARC 11554, and Utri Merah backcrossed toIR64, respectively, were developed by Angeles et al (1998). They showed resistanceto RTSV and tolerance to RTBV in endemic areas in the Philippines, India, and Indonesia(Cabunagan et a1 1995, 1998). The first two lines were resistant to RTSV butsusceptible to GLH, while the third was resistant to both RTSV and GLH. Preliminarytrials at the PhilRice Central Experiment Station (CES), Nueva Ecija, and BranchStation at Midsayap showed that IR71031-4-5-5-1 and IR69726-116-1-3 yielded 3.5-4.5 t ha -1 . IR71031-4-5-5-1 produced 3.5-4.8 t ha -1 in farmers' field trials at BualNorte, Midsayap, during the 1998 wet season. These two lines are ideal stopgap materialsfor farmers in RTD-endemic locations. Seed samples of these lines were givento farmers during the graduation of the Farmers' Field School and Workshop on RTDManagement (Natural Resources Institute [NRI]-IRRI-PhilRice-DA) in December1998. More collaborative efforts on plant breeding between IRRI and the nationalagricultural research system will be undertaken to broaden the genetic background ofdonor parents in the development of RTD-resistant lines, and to cope with the urgentneed of farmers for seeds.Seed productionThe Seed Production Program of PhilRice plays a vital role in increasing breederseeds of approved varieties. It maintains seeds of 63 recommended varieties for researchpurposes of the members of the National Rice R & D Network such as theTechnology Demonstration Farm Project (Gintong Ani, 12 Steps in Rice Production,Grain Production Enhancement Program IV, [DA 1996]). It also provides foundationseeds for the National Rice Seed Production Network (SeedNet) and sells these to4 Truong et al

seed growers and farmers to increase the registered seed requirements of the country.The area covered by SeedNet increased from 95 ha in 1994 to 354 ha in 1996. SeedNetcontinues to be an important partner of the PhilRice program to sustain grain productionand distribute registered seed requirements of seed growers and farmers all overthe country. Among the 10 preferred irrigated lowland varieties, IR62, IR74, PSB Rc10, PSB Rc 18, and PSB Rc 34 were in great demand in Mindanao because of theirtolerance for the RTD vector.PhilRice’s Communication and Training Division and Seed Production Divisionhave collaborated in coordinating training courses on seed production and producingtechnical data sheets and information on agrocharacteristics of varieties recommendedfor different rice ecosystems.Technology promotionRTD management is a part of the integrated pest management (IPM) curriculum inthe Rice Technology Demonstration (Techno Demo) Project of the Gintong Ani RiceProgram of the Department of Agriculture. This project was established to demonstratethe latest rice farming technologies in different locations with high-yieldingvarieties, nutrient and pest management, and farm machinery components. It is spearheadedby PhilRice in collaboration with other agencies of the DA, local governmentunits (LGU), state colleges and universities, and farmers’ groups. Macasieb et a1 (1996)and Javier (1999) describe the project’s technical, coordinating, and implementingprotocols, as well as the salient features of the technology package. The initial phasestarted with 1,000 technology demo farms (1 ha each) in 1995 and expanded to 1,469in 1996 and 2,940 in 1997, with the size reduced to 0.5 ha each. The project wasimplemented by farmer-cooperators chosen and supervised by municipal agriculturalofficers and technicians. The supervisors were graduates of the Season-Long RiceSpecialist Course and Training of Trainers on IPM Course. They conducted/facilitatedthe Farmers’ Field School for farmer-cooperators and other farmers. The demonstrationfarms served as venues for farmer-participants to discuss and try out newtechnologies. The FFS aims to promote the manipulation of the agroecosystem tominimize the disturbance of natural elements in the ecosystem through integratedpest management (Rola et a1 1998). From 1993 to 1996, 2,410 rice FFS were established.Survey data from 308 FFS showed that rice farmers who practiced synchronousplanting increased from 57% before the project was implemented to 75%. Theirpreferred varieties were IR60, IR64, IR74, PSB Rc 10, and PSB Rc 18.One strategic plan of the second phase of the project started in 1998 to focus onsynchronous planting as a fundamental step in pest management. This is now carriedout on 207 20-ha “compact” and eight villagewide inbred rice demonstration farms,and on eleven 20-ha hybrid rice farms nationwide. As a backup, PhilRice’s CropProtection Division, in collaboration with its Communication and Training Division,has prepared a series of simple technical materials on different components of IPM toenrich the FFS curriculum. These technical bulletins will be translated into local dialectsand made into posters by PhilRice.Rice tungro disease 5

Cultural practices in RTD managementCultural practices are the most realistic way to manage RTD because these activitiesare close to farmers’ experiences. An awareness of synchronous planting among farmersin the community is the key to the success of pest management. Synchrony can onlybe achieved with sufficient irrigation water. Most farmers in irrigated lowlands innortheastern and northwestern Luzon, where dry and wet seasons are distinct (climatictype I), have been effectively practicing synchronous planting. Planting of thefirst crop begins in December/January, followed by the nonrice crop/fallow period;the second begins in May/June. Irrigation water is provided only during the 4-moplanting period. Farmers could produce high yields even from susceptible varieties.Even in nonendemic locations in Luzon, staggered planting increased the GLH populationat later planting dates (Fig. 1A-F). Accordingly, the cumulative incidence ofRTD also increased in late planting months (Fig. lG, H). On the other hand, staggeredplanting tended to increase GLH densities earlier in the following crops, resultingin high RTD incidence (Fig. 2).Fig. 1. Population fluctuation of green leafhoppers at different planting dates in the 1994-95 dryseason (A-C) and in the 1995 wet season (PF) and the cumulative incidence of rice tungro disease(G and H for dry and wet seasons, respectively) at the PhilRice Experimental Station, Maligaya,Nueva Ecija.6 Truong et al

Fig. 2. Incidence of rice tungro disease (RTD) in varieties with different resistance characteristicsunder four methods of crop establishment. DS = direct seeding.Synchrony can also be achieved by rotating two rice crops with a nonrice cropsuch as mungbean, as farmers do in Hagonoy, Davao del Sur. Another importantcomponent of cultural practices is the broadcasting method or direct seeding. Cropestablishment by direct seeding has become popular in both dry and wet seasons inMidsayap to shorten planting time right after irrigation during land preparation. Farmersconsider this method to be easier, faster, and cheaper than transplanting (Truong et al,Community-based rice pest management, this volume). GLH-resistant varieties suchas IR56 established by transplanting or direct seeding had the lowest incidence ofRTD (2–12%) (Fig. 2A-D). A high seeding rate (120 kg ha -1 ), however, reduced theinfection rate of RTD on GLH-resistant varieties IR62 and PSB Rc 34 and susceptiblevarieties IR64 and IR66 during the first 4–5 wk after broadcasting (Fig. 2B). Diseaseincidence on IR64 reached as high as 60-80% as the crop grew older, while diseaseincidence on IR62 and PSB Rc 34 remained low (8–18%). Likewise, rotation of twoto three GLH-resistant varieties after two to three cropping seasons and synchronoustransplanting based on communal seedbed preparation were promoted by the BoholAgricultural Promotion Center (APC), and are being practiced by farmers within the40-ha contiguous irrigated farms. APC regularly monitored leafhoppers andplanthoppers by kerosene light trap, and conducted training for extension workers ofLGUs on identifying RTD symptoms and on other diagnostic tests.Rice tungro disease 7

ConclusionsRTD has declined markedly in Central Luzon rice production areas since the majoroutbreak in 1971. However, it remains the most important rice disease in the irrigatedlowland ecosystem, especially in locations with insufficient irrigation water and staggeredplanting practices such as some areas in Mindanao. These are commonly knownas “hot spots” or endemic rice tungro areas. Major outbreaks have sporadically occurredin the past 8 yr and affected from 900 to 2,700 ha annually. More research andextension efforts are needed to understand and regularly monitor the agroecosystemto forecast RTD development and to avoid disease outbreaks. RTD management inthese areas largely depended on the campaign to practice regular planting with a fallowperiod or synchronous planting, availability of GLH-resistant varieties, and insecticideapplication. The introduction of a few advanccd breeding lines resistant torice tungro virus has just begun and is limited to some farmers' groups in rice tungroendemicvillages in Midsayap, North Cotabato.ReferencesAlfonso AA, Viray MCC, Maramara GV, Tiongco ER, hfartin HC. 1906, Transfer of tungroresistance from O. rufipogon into cultivated rice. In: Philippine R & D highlights 1996.Maligaya, Muñoz, Nueva Ecija (Philippines): DA-PhilRice. p 38–39,Angeles ER, Cabunagan RC, Tiongco ER, Azzam O, Teng PS, Khush GS, Chancellor TCB.1998. Advanced breeding lines with resistance to rice tungro viruses. Int. Rice. Res. Notes23(1):17–18.Arboleda M, Sta. Cruz F, Cabauatan PQ, Muhsin M, Bhatrarai I, Baw A, Anh H, Azzam O.1997. Genomic variation of rice tungro bacilliform virus in the Philippines, Indonesia,and Vietnam. Abstract submitted at the General Meeting of The lnternational Program onRice Biotechnology, 15–19 September 1997, Malacca, Malaysia.Bajet NB, Daquioag RD, Hibino H. 1985. Enzyme-linked Immunosorbent assay to diagnosetungro. J. Plant Prot. Trop. 2:125–129.Baria AR. 1997. Status of rice tungro disease in the Philippines: a guide to current and futureresearch. In: Chancellor TCB, Thresh JM, editors. Epidemiology and management of ricetungro disease. Chatham (UK): Natural Resources Institute.Cabauatan PQ, Koganezawa H. 1994. Symptomatic strains office tungro bacilliform virus. Int.Rice Res. Newsl. 19(2):11–12.Cabunagan RC, Angeles ER, Tiongco ER, Villareal S, Azzam O, Teng PS, Khush GS, ChancellorTCB, Truong XH, Mancao S, Astika IGN, Muis A, Chowdhury AK, Ganapathy T,Subramanian N. 1998. Multilocational evaluation of promising advanced breeding linesfor resistance to rice tungro viruses. Int. Rice Res. Notes 23(1):15–17,Cabunagan RC, Angeles E, Tiongco EC, Villareal S, Truong XH, Astika IGN, Muis A,Chowdhury AK, Ganapathy T, Chancellor TCB, Teng, PS, Khush GS. 1995. Evaluationof rice germplasm for resistance to tungro disease. In: Chancellor TCB, Teng PS, HeongKL. editors. Rice tungro disease epidemiology and vector ecology. IRRI-NRI. p 92–99.Dahal G, Hibino H, Cabunagan RC, Tiongco ER, Flores ZM, Aguiero VM. 1990. Changes incultivar reactions to tungro due to changes in “virulence“ of leafhopper vector. Phytopathology80(7):659–665.8 Truong et al

De los Reyes JB, Cabunagan, RC, Coloquio E, Azzam O. 1998. Genetic variation of rice tungrobacilliform virus in four tungro-outbreak provinces in the Philippines. A poster presentedat the Workshop on Rice Tungro Disease Management, 9–11 November 1998. InternationalRice Research Institute, Los Baños, Laguna.Department of Agriculture (DA). 1996. Grains production enhancement program (GPEP). 12steps in rice production. PhilRice, Muñoz, Nueva Ecija and Information Division. DA,Quezon City. p 44.Hibino H, Roechan M, Sudarisman S. 1978. Association of two types of virus particles withpenyakit habang (tungro disease) of rice in Indonesia. Phytopathology 68:1412–1416.Hibino H, Tiongco ER, Cabunagan RC, Flores ZM. 1987. Resistance to rice tungro-associatedviruses in rice under experimental and natural conditions. Phytopathology 77:871–975.Hibino H, Ishikawa K, Omura T, Cabauatan PQ, Koganezawa H. 1991. Characteritation ofice tungro bacilliform and spherical viruses. Phytopathology 81:1130–1131.Javier LC. 1999. Gintong Ani Program: Rice Research and Development Component TechnologsDemonstration Project, Protocol 1998–99. PhilRice report (unpublished). p 80.Khush GS. 1977. Disease and insect resistance in rice. .Adv. .Agron. 29:265–341.Khush GS. 1989. Multiple disease and insect resistance for increased yield stability in rice. In:Russel IGE, editor. Progress in plant breeding. Oxford (England): Blackwell ScientificPublications. p 239–279.Macasieb ZC, Serrano SR, Malabanan FM, Javier LC, Rebuelta PI, Lara RJ, Justo HJ, FranciscoSR, 1996. Gintong ani rice technology demonstration. Philippine Rice Research andDevelopment highlights. Muñoz, Nueva Ecija (Philippines): PhilRice. p 198–199.Obenita AC. 1998. Office of Provincial Agriculturist. Pest and disease monitor in DA Bukidnon.Valencia. Bukidnon.Padolina TF. 1995. New approved rice varieties. In: Plenary paper presented during the 9thNational Rice R & D Review and Planning Workshop, 1–3 March 1995, PhilRice, Maligaya,Muñoz, Nueva Ecija.Rola AC, Provido ZS, Olanday MO, Paraguas FJ, Sirue AS, Espadon MA, Hupeda SP. 1998.Making farmers better decision-makers through the Farmer Field School. Technical Bulletin.SEAMEO Regional Center for Graduate Study and Research in Agriculture. 28 p.Savary SO, Fabellar N, Tiongco ER, Teng PS. 1993. A characterization of rice tungro epidemicsin the Philippines from historical survey data. Plant Dis. 77:376–382.Serrano FB. 1957. Rice accep na pula or stunt disease-severe menace to the Philippine riceindustry. Philipp. J. Crop Sci. 86:203–323.Tiongco ER, Chancellor TCB, Villareal S, Magbanua MGM, Teng PS. 1998. Roguing as atactical control for rice tungro virus disease. J. Plant Prot. Trops. 11:45–52.Villegas LC, Drura A, Bajet NB, Hull R. 1997. Genetic variation of rice tungro bacilliformvirus in the Philippines. Virus Genes 15(3):195–201Warburton H, Palis F, Villareal S. 1997. Farmers’ perception of rice tungro disease in the Philippines.In: Heong KL, Escalada MM, editors. Pest management of rice farmers in Asia.Manila (Philippines): International Rice Research Institute. p 1–19.Warburton H. Palis F, Villareal S, Pingali PL. 1996. Socioeconomic studies on rice tungrodisease. In: Chancellor TCB, Teng PS, Heong KL. editors. Rice tungro disease epidemiologyand vector ecology. IRRI-NRI Collaborative Project. Phase I. 1990–95. p 100–103.Yambao Ma. LM. Muhsin M, Cabauatan PQ, Azzam O. 1997. Molecular characterization oftwo rice tungro spherical virus strains and their distribution in Philippines and Indonesia.Abstract submitted to the General Meeting of The International Program on Rice Biotechnology,15–19 September 1997, Malacca. Malaysia.Rice tungro disease 9

NotesAuthors’ addresses: X.H. Truong, E.R. Tiongco, E.H. Batay-an. S.C. Mancao, Philippine RiceResearch Institute (PhilRice), Maligaya, Muñoz, Nueva Ecija, 3119; M.J.C. Du, BoholAgricultural Promotion Center/Bohol Integrated Agriculture Promotion Project, Dao District,Tagbilaran City, Bohol 6300; N.A. Juguan, Office of The Provincial Agriculturist,Province of Negros Occidental, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.10 Truong et al

Preliminary analysis of genetic variationof rice tungro bacilliform virus in twoprovinces of the PhilippinesM. Arboleda, F. Sta. Cruz, and O. AzzamA basic understanding of tungro virus populations is a prerequisite for anydeployment strategy of conventional or transgenic virus resistance. In the1996 and 1997 wet seasons, the genetic variability of rice tungro bacilliformvirus (RTBV) field populations was monitored in lsabela and NorthCotabato provinces of the Philippines. Based on restricted genome DNAprofiles and Pearson’s correlation coefficient analyses, heterogeneous anddistinct RTBV populations were identified in the two provinces. Althoughmembers of the populations reoccurred in some sites, the combination ofgenotypes differed significantly over time, suggesting a rapid evolution ofthe virus population. This study shows that changes in virus populationsneed to be continuously monitored to better understand and predict tungrooutbreaks and to prolong the life of deployed resistance genes.In highly intensive irrigated rice ecosystems in Southeast Asia. tungro disease causesconsiderable yield losses. The rice tungro disease complex is associated with ricetungro bacilliform virus and rice tungro spherical virus. On its own. RTBV causesyellowing and stunting symptoms but it cannot be transmitted by leafhoppers unlessRTSV is present. RTBV is a dsDNA belonging to the pararetrovirus group. a group ofplant DNA viruses that replicate through an RNA template. A DNA hybridizationtechnique was developed to differentiate the viral genomic DNA of four biologicalvariants of RTBV using total DNA extracts of infected plants (Cabauatan et al 1998).The technique was also applied to examine the variability of natural field populationsof RTBV in tungro-endemic areas of the Philippines. Genomic DNA profiles representingsingle infections based on molecular weight estimates were selected and comparedamong the surveyed sites.Materials and methodsRandom samples were collected from the tungro hot spot provinces of Isabela andNorth Cotabato (30–50 samples per field and 4-6 fields per province; Fig. 1 ). Thesesamples were then assayed by enzyme-linked immunosorbent assay (ELISA) againstRTBV and RTSV antisera (Cabauatan et al 1995). Following the procedure developedby Cabauatan et a1 (1998), total DNA was extracted using a modified cetyltrimethylammonium bromide (CTAB) method.The pelleted DNA was resuspended in 50 m L of sterile distilled water. Three tofive micrograms of total plant DNA was then digested with 30-50 units of EcoRVand incubated overnight at 37 °C. The samples were then electrophoresed in 0.8%agarose gel for 5 h at 100 V in 1 x TBE buffer and blotted onto Hybond-N nylonmembrane using 2 x SSC. Blots were baked at 80 °C for 2 h prior to hybridizationwith the full-length RTBV-Ic clone. Chemiluminescence ECL (Amersham) was usedfor detection following the manufacturer's instructions.

Fig. 1. Sampling sites for the study in lsabela and North Cotabato provinces in the Philippines.After hybridization, restriction fragment length polymorphisms (RFLP) of differentgenotypes were used to determine genotype frequencies across locations andtimes. A dendrogram was also generated using Pearson’s correlation coefficient todetermine the geographic distribution of the different genotypes.ResultsFollowing digestion with EcoRV of total DNA extracts from 731 isolates and DNAhybridization, 24 distinct DNA genotypes were identified from Isabela and NorthCotabato (Figs. 1 and 2).High genetic variation was observed among fields and between provinces. Geneticvariation was also significantly different over time based on the bulk Chi squareanalysis of major, minor, and mixed genotypes (Fig. 3, Table 1). Results showed that2 to 9 distinct genotypes occurred per field and a few dominated at one time. Genotypes3 and 5 were detected in all sites of Isabela and genotypes 11 and 12 wereobserved in all sites of North Cotabato. Although some genotypes were commonamong sites, each site had its unique set of genotypes and genotype frequencies. Datagathered from North Cotabato in the 1997 dry season (DS) exhibited a spatial distributionpattern similar to that of North Cotabato in the 1997 wet season (WS). Betweensites, fields sampled in Isabela in the 1996 WS were similar to each other intheir set of genotype frequencies, whereas Ilagan and Cauayan in the 1997 WS differedsignificantly in their set of genotype frequencies.12 Arboleda et al

Fig. 2. Distinct rice tungro bacilliform virus (RTBV) genotypes detected in the Philippines. Genotypes1-9 were initially found in lsabela while genotypes 10-24 were initially found in North Cotabato. Lanemarked M is a DNA size marker from the RTBV-lc3 clone.RTBV mixed infections (based on the total number of DNA fragments detected>16 kb) were common and accounted tor 10–54% of the total sample population.During the 1997 DS, their frequency was even higher than for a single infection inNorth Cotabato (Fig. 3, Table 1).Based on the frequency of the genotypes using Pearson’s correlation coefficient,RTBV genotypes were distributed geographically (Fig. 4). This showed that the genotypefrequencies between Isabela and North Cotabato were significantly different. InIsabela, 10 distinct genotypes (1, 2, 3, 5, 9, 11, 17, 16, 19, and 35) were identifiedduring the 1996 WS and in the 1997 WS. Only five genotypes (3, 5, 9, 19, and 25)were identical to those detected earlier from the same province. Genotypes 3 and 5generally dominated in the 1996 WS, but these genotypes were augmented with genotypes9 and 25 in the 1997 WS. A similar pattern was observed in North Cotabatoduring the 1997 DS and WS. In the 1997 DS, 13 distinct genotypes were observed (2,3, 10–16, 18–20, and 24) and, in the following season (1997 WS), 11 genotypes wereidentified (10–16, 18–20, and 24). Genotypes 10–13 were the most frequent in the1997 DS and remained dominant in the 1997 WS.Preliminary analysis 13

Fig. 3. Representative fields in M’lang and Pigcauayan, North Cotabato, showing the genetic variationof RTBV genotypes between planting seasons (1997 dry season and wet season; 97 DS and WS,respectively), among fields (M’lang vs Pigcauayan) and within fields. Vertical arrows indicate identicalgenotypes found in the two seasons. Slanting arrows show mixed infections. Lane marked M is aDNA size marker from the RTBV-lc3 clone.Table 1. Frequency, proportion, and x 2 for comparing the distribution of major,minor, and mixed genotypes of rice tungro bacilliform virus in North Cotabato,1997 dry and wet seasons.SeasonGenotypesMajor Minor Mixed1997 DS1997 WS10 11 12 13 514(12.0) a10(12.0)18(15.5)22(15.5)14(12.0)15(12.0)12(10.3)22(10.3)NF b 29(25.0) (25.0)10 40(7.7) (37.0)x 2 =23.17 c2911(10.2)a Numbers in parentheses are proportions.b Not found in 1997 dry season.c Frequency of typesdiffers significantly (P < 0.01) across seasons.14 Arboleda et al

Fig. 4. Dendogram depicting the relationships among rice tungro bacilliform virus (RTBV) genotypesat different sites in the Philippines. RTBV genotype frequencies were used to calculate Pearson’scorrelation coefficient between sites, and the correlation matrix was used to construct the dendogram.ConclusionsBased on the EcoRV genome profile assay, RTBV populations are genetically variablein the field. The data showed that more than one RTBV isolate could be found inone location. Even with just two planting seasons studied, results showed that RTBVpopulations exist as a diverse group of variants and are continuously evolving. Thisstudy showed that single isolates from a given location are not necessarily representativeof or specific to that location. Virus populations need to be monitored continuouslyto ensure that varietal screening for deployed rice varieties uses appropriatevirus isolates in the area.ReferencesCabauatan P, Cabunagan R, Koganezawa H. 1995. Biological variants of rice tungro viruses inthe Philippines. Phytopathology 85:77–81,Cabauatan P, Arboleda M, Azzam O. 1998. Differentiation of rice tungro bacilliform virusstrains by restriction analysis and DNA hybridization. J. Virol. Methods 76:121–126.Preliminary analysis 15

NotesAuthors’ address : M. Arboleda, E Sta. Cruz, and O. Azzam. International Rice Research Institute,MCPO Box 3127, Makati City 1271, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.16 Arboleda et al

Preliminary analysis of genetic variationof rice tungro spherical virusin the PhilippinesK.M.L. Umadhay, M.L.M. Yambao, and O. AzzamThis study investigates the genetic variability of rice tungro spherical virus(RTSV) in nine tungro-endemic sites of the Philippines and over the 1996wet season and 1997 wet and dry seasons. Based on a reverse transcriptase-polymerasechain reaction technique (RT-PCR) followed by restrictionenzyme analysis, six distinct genotypes of RTSV were identified andtheir frequency across all sites determined. Results showed that morethan one genotype could exist in a plant and at least two RTSV genotypesare present at one site. Although RTSV population did not change duringthe sampling, the presence of mixed infections and minor genotypes suggestthat the structure and composition of the virus population is not stable.it is essential to continue monitoring these populations over an extendedperiod to identify factors that lead to virus outbreaks or extinction of thecurrent prevailing populations. This approach is critical in achieving durablevirus resistance.In the tungro disease complex, rice tungro spherical virus assists in the semipersistenttransmission of rice tungro bacilliform virus (RTBV). the other component. whichcauses tungro symptoms. Earlier studies showed that rice varieties react differently toRTSV variants. Recently, polymorphic molecular markers were developed to differentiatetwo variants of RTSV, RTSV-A and RTSV-Vt6 (Yambao et al 1998). Thesevariants are serologically indistinguishable. The molecular markers are based on theamplification of coat protein regions 1 and 2 (CP1 and CP2) of the viral genomeusing reverse transcriptase-polyinerase chain reaction (RT-PCR) followed by a restrictionanalysis of the PCR product. Using this method, the genetic variation ofRTSV natural populations in tungro-endemic regions of the Philippines was investigatedin 1996 and 1997.MethodsNine tungro-endemic sites were sampled during the 1996 wet season (96 WS) and1997 dry and wet seasons (97 DS and WS). Figure 1 shows the sampling sites. Allsamples were tested by enzyme-linked immunosorbent assay (ELISA) against bothRTBV and RTSV antisera and ELISA-positive samples for RTSV were processed byKT-PCR and restriction analysis. The RT-PCR of CP regions 1 and 2 included (1)extraction of total RNA using trizol reagent. (2) cDNA synthesis using Superscript II,and (3) PCR using the cDNA as a template mixed with a cocktail of the followingreagents: Taq DNA polymerase. primers. MgC12, dNTP mixture, and buffer. The reactionconditions were initial denaturation at 95 °C for 3 min. denaturation at 95 °Cfor 1 min. annealing temperature of 50 °C for 1 min. extension of68 °C for 5 min, andfinal extension of 72 °C for 7 min. The generated PCR products were then digested

Fig. 1. Sampling sites in tungro-endemic provinces of the Philippines. Planting seasons and ricevarieties are indicated on the right side.with Hind III and Bst Y I restriction enzymes and run on 1% agarose gel stained withethidium bromide (Table 1).Results and discussionSix distinct coat protein genotypes were identified in the initial 302 analyzed samplesfrom North Cotabato, Nueva Ecija, and Bicol provinces during the 1996 WS and DS,and 1997 WS. The frequency of each coat protein genotype was used as a potentialindicator of virus variation per location (Fig. 2). In Nueva Ecija and Bicol, coat proteingenotypes II, III, and VI were observed. In North Cotabato, coat protein genotypesII, III, V, and VI were found in the 1997 DS while coat protein genotypes I, II,18 Umadhay et al

Table 1. Size characteristics of the six distinct coat proteingenotypes in the natural RTSV population of the Philippinesbased on restriction analyses of RT-PCR productswith Hind lll and Bst Y1.Coat proteingenotype numberIIIIII (RTSV-Vt6)VVIRTSV-AMixaHind lll (kb)1.151.151.150.580.580.58MixBst Y1 (kb)1.151.000.70, 0.30,0.201.151.000.80, 0.28MIXa Mix = variable and mixed enzyme restriction patterns.Fig. 2. Diversity of coat protein genotypes in the different tungro-endemic provinces of the Philippines.III, V, and VI were present in the 1997 WS (Fig. 2). In both the 1996 and 1997 croppingseasons, pattern II dominated the RTSV population in the three locations butminor genotypes and mixed infections were also observed in all locations (Table 2).Preliminary analysis 19

Table 2. Distribution of coat protein genotypes of rice tungro spherical virus based onlocation in the endemic regions of the Philippines.Location/season aNueva Ecija 96 WSBicol 96 WSNorth Cotabato 97 DSNorth Cotabato 97 WSCoat protein genotypesI II Ill v VI–––417436292155––218169Mix–617–PCR (+) Totalanalyzed27126648153511290149302a WS = wet season, DS = dry season.Table 3. Distribution of coat protein genotypes of rice tungro spherical virus based on variety in theendemic regions of the Philippines.Variety/location/seasonIR74 (M’lang 97 DS)IR74 (Bicol 96 WS)IR64 (M'lang 97 WS)IR64 (Nueva Ecija 96 WS)IR68 (Pigcauayan 97 DS)PS BRc6 (Bicol 96 WS)PS BRc8 (Bicol 96 WS)PS BRc10 (Bicol 96 WS)PS BRc14 (Bicol 96 WS)RC14 (Nueva Ecija 96 WS)I––––––––––Coat protein genotypeMix TotalII III V VI analyzed17(56.7) a–18(37.5)13(41)–––3(100)1(20)2(29)3(10)–1(2.1)2(6)––1(50)–––––––––5(10.4)8(25)–1(100)–3(10)1(100)––5(16.7)–1(50)– – – –– – –– – –4(80)–231403120123517Selection 55 (Kabacan 97 DS)Unknown (Tulunan 97 DS)Unknown (Nueva Ecija 96 WS)Farmer’s 39 (Kabacan 97 WS)Masipag (Tulunan 97 WS)Malagkit (Pigcauayan 97 WS)––––1(1.8)3(6.7)6(20)13(43.3)2(67)1(2.3)3(5.4)7(15.6)2(6.7)–––4(7.1)–2(6.7)–––1(1.8)–3(10)3(10)––1(1.8)3(6.7)5(16.7)4(13.3)––––22253313642302a Values in parentheses are frequencies of the corresponding genotypes.Mixed infections were observed mainly in North Cotabato and Bicol. Coat proteingenotype II predominated in both IRRI and local varieties. IR74 and IR64 were susceptibleto RTSV coat protein genotypes II and III. On the other hand, local varietieswere susceptible to a wide range of RTSV coat protein genotypes (Table 3).These results suggested that RTSV populations in the field are heterogeneousand more than one RTSV variant could exist in one plant, that at least two variantscould coexist at one location, and that the dominant variant may persist over timedepending on the selection pressures present in that location. During the sampling,RTSV populations did not change and one genotype seemed to dominate in the Philippinesunder current conditions. In addition, IRRI varieties were similar to other20 Umadhay et al

varieties and did not seem to exert special selection pressure on current RTSV populations.Because of the presence of mixed infections and minor genotypes, however,our data suggest that the structure and composition of the virus population may changeif a different selection pressure such as a host or environmental shift occurs. It istherefore essential to monitor virus populations when virus-resistant varieties are deployed.ReferenceYambao MLM, Cabauatan PQ, Azzam O. 1998. Differentiation of rice tungro spherical virusvariants by RT-PCR and RFLP. Int. Rice Res. Notes 23(2):22-24.NotesAuthors' address: K.M.L. Umadhay, M.L.M. Yambao, and O. Azzam. International RiceResearch Institute, MCPO Box 3127, Makati City 1271, Philippines.Citation: Chancellor TCB, Azzam O. Heong KL. editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños. Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Preliminary analysis 21

Breeding for rice tungro disease resistanceat PhilRiceL.S. Sebastian, E.R. Tiongco, D.A. Tabanao, G.V. Maramara, S. Abdula, and E.B. TabelinBreeding for resistance to rice tungro disease is a major objective of therice breeding program for irrigated lowland and rainfed ecosystems at thePhilippine Rice Research Institute. Tungro-resistant donors are identifiedusing visual screening and serological assays and are then crossed withselected high-yielding modern genotypes with good grain quality. Field evaluationof advanced breeding lines is carried out at experimental stations inNueva Ecija and in North Cotabato. Methods for generation advance andresistance screening are described. As of the 1998 wet season, several F8lines from the following crosses showed good resistance to tungro: PSBRc4 x TI-11-8, BPI Ri10 x TI-11-8, IR64 x TI-11-8. Molecular markers arecurrently being used to map resistance genes to tungro from Utri Merahand Utri Rajapan and molecular techniques are being developed to characterizestrains of rice tungro spherical virus.The development of tungro-resistant varieties remains a primary concern in the Philippinesbecause tungro is still the most destructive rice virus disease in the country.Almost all resistant varieties so far developed and released are resistant to the insectvector ( Nephottetix virescens ) only. A few varieties have resistance to rice tungrospherical virus (RTSV). Chemical control against the vector is very costly and thedisease can still occur even if the vector population is not high. Furthermore. there isno effective chemical control measure against tungro viruses. As such, the developmentof varieties resistant to this disease is a priority objective of the Philippine ricebreeding program.Rice tungro disease concerns at PhilRiceBreeding for yield and grain qualityThe breeding program for irrigated lowland (transplanted and direct-seeded) andrainfed rice ecosystems in the Philippine Rice Research Institute (PhilRice) has alwaysincluded tungro resistance as an objective. Selection for desirable plant morphologyand grain structure starts at F 3 . Segregating generations are advanced withcontinuous selection for yield, grain quality, and other important traits. Resistance toimportant insect pests and diseases is incorporated by crossing selections with highyield and good grain quality with breeding lines and traditional accessions carryingthe desired resistance traits. For tungro resistance, the most widely used materials areTI-11-8 (a BC4Fn derived from ARC11554 × TN1) and Utri Merah. To screen for ricetungro disease (RTD) resistance specifically, lines advanced from these crosses areshuttled to Midsayap, North Cotabato, where tungro incidence is consistently higheven during the dry season. Each entry is observed for visual symptoms caused by thedisease. Lines found to have some amount of resistance are included in multilocationpreliminary and advanced yield trials for further and final-stage tests. This method ofselection for tungro resistance, however, usually results in the selection of lines thatare resistant only to the vector.

Developing breeding lines with RTD resistanceIn 1994, a special breeding project began with tungro resistance as the primary trait ofinterest. The project aims to develop lines with resistance to tungro using variousdonor sources. By backcrossing and/or selection in the segregating population, resistancegenes are transferred to a modem genetic background. The project seeks toproduce breeding materials with durable resistance to either RTSV or rice tungrobacilliform virus (RTBV) or to both viruses, with or without resistance to the vector.With the availability of these breeding lines, the efforts of breeders working mainlyon high yield and good grain quality will become easier and faster. This prevents theoccurrence of undesirable recombinants that arise when landraces are used as sourcesof resistance. It also disrupts the progress in generation advance by reintroducingmany undesirable traits into lines that have been previously considered promising forother traits. Thus, instead of, for instance, using Utri Merah directly, the resistancegene can be transferred into promising selections by way of breeding lines containingthe resistance gene but with improved characteristics.For most of the activities under this project, workers at the PhilRice CentralExperimental Station (CES) in Muñoz, Nueva Ecija, work closely with researchers atthe PhilRice Midsayap Experiment Station (MES) in North Cotabato. CES acts mainlyas a source of materials while conducting field tests on its own and performs the moresophisticated laboratory procedures. MES serves as a satellite station for field testsand as a second source of leaf samples to be analyzed by enzyme-linked immunosorbentassay (ELISA), which is the responsibility of CES. Midsayap is strategically locatedbecause of the constantly high insect and disease pressure in the area notably includingtungro. With the strengthening of networks, the research group at MES has startedperforming its own crossing work and is now also equipped with a green leafhopperrearingnursery for virus strain maintenance and test tube inoculation.Breeding and screening strategiesModern varieties, landraces, introductions, and breeding lines are first evaluated fortheir response to tungro disease using visual screening and ELISA. Selected moderngenotypes are then crossed with resistance donors. Table 1 lists materials used in thecrossing work conducted from 1994 to 1996.F 1 and F 2 plants are reared in CES. Individual plant selection begins with F 2plants, which are space-planted in the field. During the wet season (WS), when diseasepressure is high, selection is primarily for resistance (visual screening). Duringthe dry season (DS), when disease pressure is insignificant, selection is primarily formorphology and grain structure.Starting at F 3 , dual-location testing is carried out at both CES and MES. Duringwet seasons, generation advance and field screening are conducted at both CES andMES. Only generation advance activities are conducted at CES during dry seasons,however, because of very low occurrence or nonoccurrence of tungro in Central Luzonat this time of the year.24 Sebastian et al

Table 1. Parental materials used in hybridization work for rice tungro disease resistance,1994-96.Resistance donorsUtri MerahModern varieties/lineslR65564-22-2-3lR65597-134-2-3-1lR65597-17-4-3-3lR65598-112-2lR65600-12-3lR65600-27-1-2-2lR65600-32-4-6-1lR65605-6-2-3-2lR66158-38-3-2-1lR66159-164-5-3-5lR66160-5-2-3-2lR66165-24-6-3-2lR66738-118-1-2lR66750-6-2-1IR66764-AC4-8lR66160-121-4-1-1ARC11554T1-11-8T15725T15850T15943T15950T15991T16000T16270T16291T16393LS519LS551TK298TK300TK303TK342TK352DS1-1DS1-7KataribhogUtri RajapanLS546IR22M seriesBPI Ri10C4-63GIR24IR56IR64IR72PSB Rc4PSB Rc14PSB Rc40lR54883IRBB5IRBB21PSB Rc40HabatakiTakanarlTodorikiwasiCR2CR3CR5CR93CR94LH422LX37LX76LX77LX93LX99LX136LX144aLX144bLX163PR26494The F 3 materials are planted in small (2.2-2.8 × 1.2-1.6 m), unreplicated plots.Selection is done within plots and seeds are bulked in each plot to constitute the nextgeneration. At F 4 and F 5 , the small plots are replicated thrice. Selection is still withinplots and seeds are also bulked, Lines that are consistently damaged by tungro in allthree replicates are dropped, whereas lines that show a high degree of resistance in allthree replicates are considered as the best selections. At F 6 through F 8 , lines are plantedin unreplicated but bigger plots (5.0 × 5.0 m) to minimize escape and/or preference ofinsects, which is especially the case when many genotypes are arrayed in successionin the field. At this stage, selections within each plot can be separated into distinctsublines, but each subline can still be composed of several bulked plants. Visual screeningis also coupled with ELISA at this stage. Promising selections are then tested inpreliminary yield trials.The ELISA technique is carried out as outlined by Bajet et al (1985). Leaves aresampled at 30 and 60 d after transplanting (DAT). Absorbance is set at 405 nm in aMicroELISA reader. For the field screening. spreader rows consisting of IR64 or PSBRc14 are planted 1 mo ahead to ensure a sufficient source of virus. Disease incidence(% infection) is observed at 30 and 60 DAT.Rice tungro disease observation nurseryThis part of the project aims to (1) evaluate selected rice collections and advancedlines for resistance to rice tungro disease and (2) identify potential parents for hybridizationwork. An initial screening of 56 commercial varieties and breeding lines wasconducted at CES and MES in the 1995 WS. The entries were assigned to 1.0 × 2.5-m plots each in three replications. Visual scoring of tungro incidence and leaf samplingfor ELISA were done at 30 and 60 DAT.Breeding for rice tungro disease 25

In CES, 28 entries had low tungro incidence (0–23%). while in MES all entrieshad high tungro incidence (47-100%).In both locations, TI-11-8 showed low virus titers of combined RTBV and RTSV(0%), RTBV (8% in CES, 0% in MES), and RTSV (0% in CES, 13% in MES), indicatinga high level of resistance to tungro. In CES, IR74, PSB Rcl8, and PSB Rc34showed low infections with combined RTBV and RTSV (0–4%), with RTBV (0-4%), and with RTSV (0–25%). In MES, LS519, LS551, TK298, and TK300 had lowcombined RTBV-RTSV (8-25%), RTBV (4–25%). and RTSV (4–13%) infection.In 1996, the observation nursery in CES handled 399 entries, most of which wereF4 lines. Sixty-seven lines showed a high level of resistance to tungro. In MES, onlyone entry, IR69705-1-1-3-2-1, exhibited high resistance against tungro out of morethan 900 entries evaluated, the bulk of which were F 4 and F 3 lines.Generating RTD-resistant advanced linesThis task involves the following procedures: (1) advancing early generations of crossesbetween tungro-resistant genotypes and selected rice cultivars, (2) screening for RTDresistance in early and advanced generations, and (3) purifying and increasing seed ofselected lines that are resistant to RTD. Figure 1 shows the scheme for generationadvance and resistance screening.Hybridization work began in 1994, producing 14 crosses. In the succeeding years,many materials were handled in both the dry and wet seasons. In 1995, 70 crosses,256 F 1 plants from 42 crosses, and 17 F 2 populations were advanced. In 1996, 116crosses, 270 F 1 plants from 42 crosses, 16 F 2 populations. 1.047 F 3 lines from 24crosses, and 387 F 4 lines from 4 crosses were advanced. In 1997, 14 crosses, 18 F 2populations, 334 F 3 lines from 8 crosses, 464 F 4 lines from 16 crosses, 262 F 5 linesfrom 14 crosses, and 46 F 6 lines from 4 crosses were advanced. In 1998, 6 F 3 crosses,64 F 4 lines from 8 crosses, 226 F 5 lines from 8 crosses, 276 F 6 lines from 16 crosses,198 F 7 lines from 14 crosses, and 24 F 8 lines from 4 crosses were advanced.As of the 1998 WS, several F 8 lines from four crosses have been considered asthe best tungro-resistant selections. These include four lines from PSB Rc4 × TI-11-8, three lines from BPI Ri10 × TI-11-8, and one line from IR64 × TI-11-8. Table 2shows the other crosses with promising lines.BiotechnologyIn support of breeding activities, molecular markers are currently being used to mapgenes with resistance to tungro from Utri Merah and Utri Rajapan. The molecularmarkers being used are restriction fragment length polymorphism (RFLP), amplifiedfragment length polymorphism (AFLP), and randomly amplified polymorphic DNA(RAPD). Resistance from ARC11554 was mapped earlier (Fig. 2, Sebastian et a11996a, b). Molecular markers tightly linked to the RTSV resistance gene are nowbeing tested for possible use in marker-aided selection (MAS). Cloning of the RTSVresistance gene is also under way (G.O. Romero, personal communication, 1999).26 Sebastian et al

Fig. 1. Scheme for generation advance and screening for tungro resistance.Furthermore, molecular techniques are also being tested in differentiating RTSVstrains in Muñoz, Nueva Ecija, and Midsayap, North Cotabato. This study uses eightdifferential varieties and detects variation in RTSV strains using the reverse transcriptase-polymerasechain reaction technique.Breeding for rice tungro disease 27

Table 2. Best rice tungro disease-resistant selections (as of 1998wet season).GenerationF 8F 7F 6F 5PedigreePSB Rc4BPI Ri10IR64TI-11-8TI-11-8TK298TK298TK303TK352IR22M-2LS519LS519KataribhogLS546LS551lR66158-38-3-2-1lR66738-118-1-2xxxxxxxxxxxxxxxxxTI-11-8TI-11-8TI-11-8IR56LX-37IR64TI-11-8TI-11-8TI-11-8TI-11-8IR64PSB Rc4PSB Rc4PSB Rc4IR64TI-11-8TI-11-8Lines431313111236191013Fig. 2. Molecular map indicating the arrangement of the green leafhopper (GLH) and rice tungrospherical virus (RTSV) resistance genes relative to the molecular markers on chromosome 4 in TNI× ARC11554 F 2 mapping population, derived using Mapmaker/Exp. 3.0 at LOD 3.0. OP 246 is aRAPD marker; CDO and RZ are Cornell markers; Y, C, and G are Rice Genome Program of Japanmarkers.28 Sebastian et al

Tungro resistance has also been introgressed into IR64 from Oryza rufipogon aspart of a wide hybridization project aiming to transfer insect and disease resistancegenes from wild species. Several BC 4 lines with resistance to RTSV and RTBV havealready been generated.Future outlookThe development of varieties resistant to tungro will remain a challenge to variousbreeding programs because of the complexity of the disease. It is hoped. however,that, with new tools and knowledge from concerted research and development efforts,the development of tungro-resistant varieties will become a reality.In PhilRice, promising F 8 lines will be evaluated in preliminary yield trials forgrain yield potential and yield component characteristics. Even materials with belowoptimumyields will be maintained, purified, and characterized as long as they haveresistance to tungro. Backcrossing to the modern parent may be done to recover asmany desirable traits as possible while maintaining the resistance to tungro at thesame time.The development of a marker-aided selection procedure for tungro resistanceusing the resistance gene from ARC11554 will continue to be undertaken becauseMAS, as a technique, has become an integral part of other breeding programs for pestand disease resistance. Molecular mapping for resistance genes and characterizationof RTSV strains constitute the other thrusts in RTD research, both of which are relevantto the development of strategies for resistance breeding.ReferencesBajet NB, Daquioag RD, Hibino H. 1985. Enzyme-linked immunosorbent assay to diagnoserice tungro. J. Plant Prot. Trop. 2:125–129.Sebastian LS, Ikeda R, Huang N, Imbe T, Coffman WR, McCouch SR. 1996a. Molecular mappingof resistance to rice tungro spherical virus (RTSV) and green leafhopper (GLH) inrice. Phytopathology 86(1):25–30.Sebastian LS, Ikeda R, Huang N, Imbe T. Coffman WR, Yano M, Sasaki T, McCouch SR.1996b. Genetic mapping of resistance to rice tungro spherical virus (RTSV) and greenleafhopper (GLH) in ARCl1554. In: Khush GS, editor. Rice genetics III. Proceedings ofthe 3rd lnternational Rice Genetics Symposium, Manila. Philippines. p 560–563.NotesAuthors’ address : L.S. Sebastian, E.R. Tiongco, D.A. Tabanao, and G.V. Maramara, PhilippineRice Research Institute (PhilRice), Central Experiment Station, Muñoz, Nueva Ecija,Philippines; S. Abdula and E.B. Tabelin, PhilRice, Midsayap Experiment Station, NorthCotabato. Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Breeding for rice tungro disease 29

Breeding for rice tungro virus resistancein IndonesiaA.A. Daradjat, N. Widiarta, and A. HasanuddinBreeding for rice tungro resistance is one of the major objectives of therice breeding program in Indonesia. Early studies were directed towarddeveloping rice varieties with good plant type, high yield, and resistance tothe green leafhopper (GLH) vector. In recent work, the breeding objectiveswere redefined to consider two additional traits: grain quality and resistanceto tungro viruses. A vigorous hybridization program involving severalcultivars with high yield, good plant type, excellent grain quality, and resistanceto tungro viruses was implemented. From the initial work, severalhigh-yielding rice varieties with resistance to GLH have been released. Preliminaryresults from this study indicated that 2,296 accessions have strongresistance to tungro viruses. Based on the range of infection rates withtungro on single-cross populations, it was observed that Utri Merah, TjempoKijik, Seratus Hari T36, and M1085c-10-1 were effective donors of tungroresistance. Membramo was the best combiner of the donor cultivars, withhigh yield and excellent grain quality. The reaction of advanced breedinglines to tungro infection varied with disease pressure and vector populationin the area.Rice tungro disease caused by rice tungro spherical virus (RTSV) and rice tungrobacilliform virus (RTBV) results in considerable losses in rice production in someirrigated ecosystems in Indonesia. Between 1968 and 1984, the disease damaged anestimated 199,000 ha of rice (Manwan et al 1985).In 1995, 12,340 ha of rice in Surakarta regency, Central Java, were severelyinfected, causing yield losses of about US$1.87 million (Anonymous 1995). Continuousand staggered planting of susceptible cultivars such as Cisadane and IR64and climatic conditions favorable for both the leafhopper vector and the disease todevelop were among the factors that favored the epidemics.Improved crop production technology that consists of improved cultivars, appropriatecultural practices, and suitable pest management is expected to reduce losses.In pest outbreaks, the use of resistant cultivars was observed to be the most effectivecontrol measure in Indonesian ecosystems. Thus, breeding for resistance to pestsand diseases was included as one of the main activities in the breeding program. Thispaper briefly reviews the efforts that have been made in developing rice cultivarsresistant to tungro disease in Indonesia.Early activitiesEarly resistance breeding work at the Central Research Institute for Food Crops identifiedtraits associated with virus resistance, such as growth habit, yield, and insectvector resistance.It was confirmed that tungro disease is transmitted by the green leafhopper( Nephotettix virescens ). This information led to the adoption of rice cultivars resistant

to green leafhopper as a strategy for reducing tungro disease incidence. Therefore,efforts to reduce losses due to tungro disease focused on developing rice varietiesresistant to GLH.Screening methodThe effects of the vector on tungro incidence were studied by field screening in LanrangExperimental Farm, South Sulawesi. Twenty-one- or 25-day-old seedlings were exposedto natural infection by tungro through GLH infestation in the field. Seedlingswere transplanted as two 10-hill rows at a spacing of 30 × 20 cm. To ensure adequateinoculum pressure, single rows of the susceptible check (TN1) were transplantedevery 10 rows and in the surrounding field 2 wk before transplanting test seedlings.Land preparation, fertilizer application, and hand weeding were done as recommended.No pesticides were applied at any stage of plant growth.Screening resultsUp to 1986, 47, 503 genotypes had been evaluated. Of these. 6,864 lines were classifiedas resistant and 8,432 as moderately resistant (Table 1). Table 2 lists some of thepromising lines resistant to tungro disease. Selections from crosses involving thesepromising lines have been released as new improved cultivars (Table 3).The data in Table 3 indicate that, if a single resistant cultivar of rice is growncontinuously in a particular area where there is year-round irrigation, the plant resistancelevel will be reduced due to the adaptation of the GLH population to the host.Continuous cropping of such GLH-resistant cultivars could increase the selectionTable 1. Rice breeding lines with resistance, moderate resistance, moderate susceptibility,and susceptibility to tungro disease in Indonesia, 1974-86.YearLines tested Plant reaction a Lines failed(no.)(no.)R MR MS S1974-751975-761976-771977-781978-791979-801980-811981-821982-831983-841984-851985-86Total4,8231,9071,6647,6224,3144,1916,0834,0275,5851,2722,7532,81247,053975335788143218111,4382331,303145913216,8647852262801,3675529541,5619311,1451151703468,4328842883451,3891,1241,7371,6821,1501,1818535443910,6851,4584158252,7471,7186891,4027141,8861831,2211,25814,5167216431361,30559900999717449174486,583a R = resistant, MR = moderately resistant, MS = moderately susceptible, S = susceptible, Source:Anonymous 1988.32 Daradjat et al

Table 2. Promising lines resistant to tungro disease in Indonesia, 1974-84.Pedigree Crosses Source of resistanceB3844-17C-SM-64-2B4076D-PN-20-10B4076D-PN-114-46B4076D-PN-167-63B41080-PN-210-40B4140C-SM-162-2B4176B-2-IRRI-MR-1B4180B-22-IRRI-MR-2B4183B-51-IRRI-MR-4B4183B-51-IRRI-MR-6B4196C-IRRI-MR-46B4196C-IRRI-MR-47Source: Suprihatno 1985.CR94-13/Pehta l-1// B5436/Pelita I-1 CR94-13, Ptb18, Ptb21lR3351-38-3-1/lR36TKM6, Ptb18, Ptb21lR3351-38-3-1/lR36TKM6, Ptb18, Ptb21lR3351-38-3-1/lR36TKM6, Ptb18, Ptb21B2484B-2-PN-29/IR40TKM6B3063/PI-1*2//IR36TKM6IR36/IR2071//B295JTKM6IR36//B459B/Paedai Ngulahi TKM6IR36/PI-1//lR4744/PI-1TKM6IR36/PI-1//IR4744/Pl-1TKM6B28508-S1-22//C4-63/Ase Bakk TKM6, Ptb18, Ptb21,GP15B2850B-S1-22//C4-63/Ase Bakk TKM6, Ptb18, Ptb21,GP15Table 3. Rice cultivars resistant to tungro disease released inIndonesia between 1972 and 1984.Reaction against tungroCultivars Year of disease within cropping period arelease1974-78 1979-82 1983-84SerayuSemeruCisadaneCipunegaraBaritoKrueng AcehSadangBahbolonCitanduyKelaraIR26IR36IR42IR5419781980198019801981198119831983198319831975197719801981MS–––––––––S bR––RMSSSSSMRR––MRMS bS bRRMRSSSSSMRSRMSSSMS ba R = reststant, MR = moderately resistant, MS = moderately susceptible, S =susceptible. b The dominant rice cultivar planted within the period. Source:Suprihatno 1985.pressure on GLH and lead to the development of a new GLH “biotype.” Consequently,the need to rapidly develop and multiply rice cultivars resistant to newly evolvingbiotypes of GLH is urgent.Breeding for rice tungro virus 33

Moreover, because a few viruliferous GLH could efficiently transmit tungro disease,it is important that cultivars be resistant to the causal viruses. The strategy ofdeveloping virus-resistant cultivars was promoted by identifying several sources ofresistance to the disease (Shahjahan et al 1990, IRRI 1996).Current activitiesCurrently, several rice cultivars are resistant to GLH, but these are mostly unacceptableto farmers due to their unsatisfactory eating quality. Thus, our breeding objectivesemphasized the incorporation of resistance, good plant type, and excellent grainquality to existing GLH-resistant cultivars. Accordingly. breeders at the ResearchInstitute for Rice (RlR) in Sukamandi started a program to breed such cultivars.Crosses were made between several cultivars with high yield, good plant type,and excellent grain quality and those with resistance to tungro virus (Table 4). Populationdevelopment was done by single, double, and multiple crossing and backcrossing.Screening methodField screening for tungro resistance was conducted in the tungro-endemic area ofTanjungsiang, about 60 km from the Sukamandi Experimental Farm.To increase the tungro disease pressure on the breeding materials, 15-day-oldseedlings seeded in plastic trays were exposed to viruliferous GLH at the rate of twoadult vectors per seedling for 24-h inoculation feeding. After 3 wk of exposure, asingle inoculated seedling was transplanted in the field in two 15-hill rows for eachline at a spacing of 20 × 15 cm.Resistant and susceptible parents were planted every 20 rows, and TN1 was establishedin surrounding plots. Standard agronomic practices were followed, includingapplication of nitrogen, phosphorus, and potassium fertilizer at the rate of 90 kg Nha -1 , 60 kg P 2 O 5 ha -1 , and 30 kg K 2 O ha -1 . Hand weeding was done as necessary, andno pesticide was applied at any stage of plant growth.Six and 8 wk after transplanting, visual scoring was conducted on individualhills by using the scoring method of Hasanuddin et al (1988) where1 = no symptoms3 = 1-10% plant height reduction with no distinct leaf discoloration5 = 11-30% plant height reduction with no distinct leaf discoloration7 = 31-50% plant height reduction and/or yellow to orange leaf discoloration9 = more than 50% plant height reduction and yellow to orange leaf discolorationPercent infection per plant was also assessed visually.Specifically for F 2 plants at 6 wk after transplanting, leaf samples of healthyplants were tested for the presence of virus particles by enzyme-linked immunosorbentassay. Only plants that showed relatively low concentrations of RTSV and/or RTBVwere used in further population development.34 Daradjat et al

Table 4. Rice cultivars used as hybridization parents in breeding for tungro resistance, Indonesia,1997-98.CultivarsDonor forSeed source Resistance to Amylose YieldIRTN96 RTBV RTSV RTV/GLH (%)Utri Merah-1 (Acc. 16680)Utri Merah-2 (Acc. 16682)Utri Rajapan (Acc. 16684)ARC 1154 (Acc. 21473)ARC 10312 (Acc. 124281)ARC 12596 (Acc. 22176)ARC 7140ARC 10343 (Acc. 12337)Shuli 2 (Acc. 26527)Seratus Hari T36 (Acc. 5346)Tjempo Kijik (Acc. 16602)IR68IR72IR74M1085c-10-124252610018192060232829689BSxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx19CEA 1ITA 323PR40-1-2-1(87)lR59682-132-1-1-2lR63872-14-2-2-1CT9162-12-8-10-1lR60819-31-1-3-2IR64S969B-265-1-4-1S3054-2D-12-2MembramoMarosBengawan SololR66160-121-4-5-3Taichung Sen 10Taichung Sen Glu 1IURON9513336436898143BS96BS96BS96BS96BS96BS96OBS96PYT96PYT96xx27xxxx1722212323231722192212xxxxxxxxxxxxxxxxxxxxxxxxIRTN96 = lnternational Rice Tungro Nursery 1996, IURON95 = lnternational Upland Rice Observational Nursery1995, xx = indicator for the presence of resistance or yield potential.Results of screeningA large population of segregating materials derived from single, double, or multiplecrosses was screened. Plants that showed resistant reactions (score: 1–3) and had goodplant type were selected.Based on the range of infection rates with tungro viruses on single-cross populations,Utri Merah, Tjempo Kijik, Seratus Hari T36, and M1085c-10-1 were effectivedonors for tungro resistance. In contrast. Membramo was the best combiner of theBreeding for rice tungro virus 35

donor cultivars for high yield and excellent grain quality. In general, all crosses producedoffspring with some degree of resistance to tungro, but most of them had poorplant type and a high level of sterility. To overcome the deficiencies of these primarycrosses, double and multiple crosses were attempted. Preliminary results of the presentstudy indicated that 2,296 selections have strong resistance to virus based on thescreening of more than 143 populations of F 3 single crosses, 200 populations of F 2double crosses, and 112 populations of F 1 multiple crosses.Additional breeding populations were also developed during the 1997-98 wetseason by backcrossing selected F 2 plants that exhibited resistance with recurrentparents having good plant type and excellent grain quality. The F 2 BC 1 progenies werescreened for tungro resistance in the 1998-99 wet season before making the nextbackcrosses.Comparative reaction of RTD incidence on breeding materialsThe first RIR breeding work on tungro screening was done during the 1997 dry seasonin Celuk, Bali, where IRRI advanced breeding lines were screened. Mass seedlinginoculation was done by researchers from the Celuk branch of the Agency forPlant Protection and Pest Assessment. Tungro symptoms in the field trial were assessedvisually. Plants that exhibited a resistant reaction and had good plant type wereharvested. Resistant selections from this particular nursery and some lines from theIRRI nursery were retested in Tanjungsiang, Subang, during the 1997-98 wet seasonand the 1998 dry season.Data showed that tungro incidence vaned among selections and locations. InCeluk, where IR64 was moderately infected by rice tungro disease (60%), some selectionshad very low infection rates. In contrast, the same selections showed highinfection rates when grown in Tanjungsiang. These results indicated that field resistanceto tungro in Celuk is not effective in Tanjungsiang, and confirmed that tungroincidence varies with inoculum pressure and the vector population prevailing in thearea. In the 1998 dry season Tanjungsiang nursery, selections were made from themore promising lines. It is hoped that these selections will result in lines homozygousfor resistance to rice tungro disease and for other major characters. Observation nurseriesof the selected lines are also being planted to determine their yield potential.ReferencesAnonymous. 1988. Laporan Tahunan 1986/1987 Balai Penelitian Tanaman Pangan Maros. BadanPenelitian dan Pengembangan Pertanian, Departemen Pertanian. 64 p.Anonymous. 1995. Tungro dan wereng hijau. Direktorat Bina Perlindungan Tanaman. 194 p.Hasanuddin A, Daquioag RD, Hibino H. 1988. A method for scoring resistance to tungro (RTV):Int. Rice Res. Newsl. 13(6):13–14.Manwan I, Sama S, Risvi SA. 1985. Use of varietal rotation in the management of rice tungrodisease in Indonesia. Indon. Agric. Res. Dev. J. 7:43–48.IRRI. 1996. The Twentieth International Rice Tungro Nursey (IRTN 1996). INGER.36 Daradjat et al

Shahjahan MB, Jalani S, Zakri AH, Imbe T. Othman O. 1990. Inheritance of tolerance of ricetungro bacilliform virus (RTBV) in rice ( Oryza sativa L.). Theor. Appl. Genet. 8:513-517.Suprihatno B. 1985. Pewarisan Sifat Ketahanan Varietas Terhadap Penyakit Tungro. In: Tungro.Risalah temu lapang pengendalian penyakit tungro di daerah Banyumas. Jawa Tengah,18-19 September 1985. p 26-32.NotesAuthors’ address: A.A. Daradjat, N. Widiarta. and A. Hasanuddin, Research Institute for Rice,J1. Raya IX Sukamandi 41256 Subang. West Java. Indonesia.Acknowledgments: We appreciate the help of Dr. M.D. Moentono and Dr. Suparyono in reviewingthe manuscript.Citation: Chancellor TCB, Azzam O, Heong KL. editors. 1999. Rice tungro disease management,Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Breeding for rice tungro virus 37

Genetic engineering of ricefor tungro resistanceO. Azzam, A. Klöti, F. Sta. Cruz, J. Fütterer, E.L. Coloquio, I. Potrykus, and R. HullGenes encoding sense and antisense viral coat proteins, polymerases,and proteases have been successfully used to engineer resistance to severalplant viruses. In this study, viral genes of rice tungro bacilliform virusand the coat protein 3 of rice tungro spherical virus were used to engineerresistance in rice against tungro infection. Rice varieties such as IR64,TN1, Taipei 309, and Kinuhikari were successfully transformed and fertiletransgenic plants were evaluated at T1 and T2 generations for their abilityto confer protection against tungro infection using insect inoculation assays.Unfortunately, none of the 71 transgenic lines tested provided protectionagainst tungro infection. Possible factors for the lack of protectionare discussed.Genetic engineering approaches expand the gene pool from which new and novelvirus resistance genes can be selected. For complex diseases of rice, such as tungro,these approaches offer two advantages: (1) the ability to transfer single genes withoutany linkage to undesirable traits, and (2) the ability to introduce novel genes that havenot been explored before in nature and that have potential to increase the durability ofresistance. Rice tungro disease incidence is unpredictable, but when it occurs, it cancause catastrophic yield losses in farmers’ communities in the irrigated rice ecosystem.For the last 15 yr, several institutions have invested substantial research effortsin studying the molecular biology of the two viruses that cause tungro, rice tungrobacilliform virus (RTBV) and rice tungro spherical virus (RTSV), and to geneticallyengineer tungro resistance in rice. In this study, we report on the resistance tests doneusing insect inoculation assays to evaluate several of these antiviral strategies designedagainst both RTBV and RTSV. The transgenic rice plants were produced at theInstitute of Plant Sciences, ETH, Zurich, Switzerland and John Innes Centre (JIC),Norwich. England, and the evaluation was done at the transgenic CL4 greenhousefacility at IRRI.Materials and methodsTables 1 and 2 describe the first 19 transgenic lines from ETH and 20 lines from JIC,respectively. Some additional 32 lines from ETH carrying the antisense RNA constructsof RTBV ORF4 were also evaluated. Seeds from each transgenic line, positivecontrols (inoculated nontransgenic plants). and negative controls (uninoculatedtransgenic plants) were sown in sterile soil and seedlings were grown in the CL4facility at IRRI. At 7–11 d after sowing (DAS), seedlings were inoculated with bothtungro viruses by insect feeding using viruliferous green leafhoppers (3–5 insectsseedling -1 ). Inoculated plants were then monitored for symptom expression and assayedfor the presence or absence of virus particles at 20 and 40 d postinoculation(DPI). For the evaluation of the first 19 transgenic lines from ETH, two sources of

Table 1. Initial 19 transgenic lines from the Institute of Plant Sciences, ETH, Zurich, Switzerland.Construct line Promoter/description Transformed Integratednumbers of RTBV sequence variety copies (no.)5.338.238.71.31.61.181.272.3127.107.640.58.210.411.811.1142.1K42.1K42.1.1K44.10.933.135S/ORF1RTBV/ORF135S/ORF3 (long coat proteinincluding Cys-His motif)35S/short coat protein(without Cys-His motif)RTBV/short coat protein(without Cys-His motif)35S/polymerase with zinc fingerRTBV/polymerase with zinc finger35S/polyrnerase35S/protease35S/RNase HRTBV/5'-half of ORF 4RTBV/3'-half of ORF 4RTBV/GUS geneTP 309TP 309TP 309TP 309TP 309TP 309TP 309TP 309TP 309TP 309TP 309KinuhikariKinuhikariTP 3091111-2>31-21-2>31313-512-3>6>41>3>3Table 2. Transgenic lines developed at John lnnes Center, Norwich, England.Line Promoter/intron/description of the viral gene Variety GenerationtransformedtestedIRI 1IR2IRI 9IRI 28IRI 37CaMV 35S/RTBV antisense within tRNAbinding site (RTBV nt # 7998-550)IR64T1/T2IRK 41CaMV 35S/RTBV antisense within tRNAbinding site (RTBV nt # 7998-550) coupledwith ribozymeIR64T1/T2IRW 1IRW 2IRW 3CaMV 35S/ubiquitin intron/RTBV antisense within tRNA binding site(RTBV nt # 7998-550) coupled with ribozymeIR64T1RH 58Ubiquitin promoter/RTSV coat protein 3Taipei 309T1IRU 1IRU 2IRU 4CaMV 35S/ubiquitin intron/RTSV coat protein 3IR64T1TNVP 1TNVP 3TNVP 4TNVP 6TNVP 7CaMV 35S/ubiquitin intron/RTBV antisensewithin tRNA binding site (RTBV nt # 7998-550)and ubiquitin/ubiquitin intron/RTSV coat protein 3TN1T1TNVO 1TNVO 3Ubiquitin/ubiquitin intron//RTBV antisensewithin tRNA binding site (RTBV nt # 7998-550)and CaMV 35S/ubiquitin intron/RTSV coat protein 3TN1T140 Azzam et al

virus inocula were used. For later experiments, only the greenhouse virus inoculumwas used.ResultsNone of the initial 19 transgenic lines, which used either the greenhouse virus sourceor a locally collected virus source from Famy, 40 km northeast of Los Baños, showedresistance to either RTBV or RTSV (Tables 3 and 4). In addition, most of the inoculatedplants showed severe symptoms such as stunted growth and leaf discoloration at20 DPI, and their viral coat protein titers, as measured by the enzyme-linkedimmunosorbent assay (ELISA), were comparable with those titers from thenontransgenic control plants. Titers varied among individual plants from differentTable 3. Percent infection of transgenic lines based on ELISA and visual scores(SS) using mylar test tube inoculation with the greenhouse isolate as inoculum.ELISA results are shown for 20 and 40 days postinoculation (DPI).Percent InfectionVisualConstruct line Plants tested 20 DPI 40 DPI scores anumbers(no.)RTBV RTSV RTBV RTSV (SS)5.338.238.72028281001001001001001001001001001001001007771.31.61.181.27282828281001001009696971001001001001009610010010010077772.3127.12828100100921001001009393777.640.52827961001008996100100100778.210.411.811.112528282610010010010010010010089100100100100100969692777642.1K42.1K42.1.1K44.10.9272827201001001001001009610010010010010010093939696777833.1 (construct alone)TP 309 (nontransgenic)Kinuhtkari (nontransgenic)2820281009510092100100100951008510096777TP 309 (uninoculated) 19 0Kinuhikari (uninoculated) 28 000000011a IRRI Standard Evaluation System (SES), 1=resistant. 9=susceptible.Genetic engineering of rice 41

Table 4. Percent infection of transgenic lines based on ELSA and visual scores(SS) using mylar test tube inoculation with Famy isolate as inoculum. ELISAresults are shown for 20 and 40 days postinoculation (DPI).Percent lnfectionConstruct. line Plants tested 20 DPI 40 DPI Visualnumbers(no.)scores aRTBV RTSV RTBV RTSV (SS)5.3 14 100 100 100 100 738.2 28 100 100 100 100 838.7 28 100 100 100 92 71.3 27 100 96 100 891.6 26 100 96 99 961.18 28 100 100 100 1001.27 28 100 96 100 96992.3128 100 100 100a IRRI Standard Evaluation System (SES), 1=resistant, 9=susceptible27.17.640.58.210.411.811.1142.1K42.1K42.1.1K44.10.9282828212828426282822100100100100100100100100100100951001009610096100100100100939110010010010010010010010010010010010010097100100100100100100968533.1 (construct alone) 27 100 92 100 100TP 309 (nontransgenic) 22 100 100 100 97Kinuhikari (nontransgenic) 28 100 100 100 100TP 309 (uninoculated)19 0 0 0 0Kinuhikari (uninoculated) 28 0 0 0 0777777878777777777711lines. Some plants had high RTSV and RTBV titers while others, surprisingly, hadlow RTSV but high RTBV titers. Based on the ELISA results and visual scores, noneof the test lines recovered at 40 DPI. The average symptom severity (SS) was about 7per line, indicating that most individual plants within a line exhibited stunted growthand leaf discoloration. The 32 remaining lines from ETH were tested using only thegreenhouse virus population and results were similar to those obtained earlier. Noneof the lines showed resistance to RTBV at 20 DPI and plants did not recover after 40DPI (data not presented).Furthermore, none of the lines from JIC showed any promising protection againsteither RTBV or RTSV. Based on ELISA and visual scores, the plants accumulatedRTBV at a level similar to that of the nontransgenic control plants. Generation T1 andT2 plants of the IRI and IRK lines did not show any protection against virus infection(Table 5).42 Azzam et al

Table 5. Percent infection of transgenic lines based on ELlSA and visual scores(SS) using mylar test tube inoculation with the greenhouse isolate as inoculum. ELlSAresults are shown for 21 days postinoculation (DPI).Line Clones tested line -1 Plants tested % infection % lnfection(no.) with RTBV with RTSVIRI 1IRI 2IRI 9IRI 28IRI 37IRK 41IRW 1IRW 2IRW 3IR64 inoculated controlTN1 inoculated controlIR64 uninoculatedRH 58Taipei 309 inoculatedTaipei 309 uninoculatedIRU 1IRU 2IRU 4TNVP 1TNVP 3TNVP 4TNVP 6TNVP 7TNVO 1TNVO 3TN1 inoculated controlTN1 uninoculated411216542Nontransgenic3322111121127383754405817927727402586392081806339374039322234402064909289100609597979610009597095851001001001001009810097100011507865100196076747410008774069636231971009794100771000DiscussionFour rice varieties, IR64, TNl, Taipei 309, and Kinuhikari, were successfully transformedwith RTBV and RTSV gene constructs and fertile transgenic plants were generated.The coat protein, polymerase, protease, RNase H, and antisense RNA resistancestrategies were used to confer protection against RTBV infection using the cauliflower35S and RTBV promoters. The coat protein strategy was tried against RTSVinfection using the 35S and ubiquitin promoters. Most of these strategies have beensuccessful in other virus systems and they were expected to be effective with DNAand RNA viruses. Unfortunately, the resistance tests showed that none of these strategieswas effective against tungro infection.In the transformation experiments with the coat protein, polymerase, and proteasestrategies, most constructs were expected to express viral proteins when integratedin rice. Most transgenic plants, however, expressed the transgenes only at aGenetic engineering of rice 43

very low level. In fact, transgene expression was either stopped or only expressed ina subset of cells. Such an irregular expression level could be responsible for the lackof protection. Novel transgene-expression strategies were thus designed and newlygenerated transgenic plants will be evaluated in the near future.Another possible factor that could be responsible for the lack of protection is thequasispecies behavior of tungro viruses (Villegas et al 1996, Cabauatan et al 1999).The continuous supply of mutant and recombinant genomes during virus replicationmay permit great virus adaptability in overcoming selection pressures imposed on itsreplication or movement within a very short time. Our work on the genetic variationof RTBV and RTSV field populations (see Arboleda et a1 and Umadhay et al, thisvolume; Azzam et al 1999) suggests that effective protection mechanisms must bedirected against highly conserved functions or sequences, which can be defined onlyby analyses of large numbers of field isolates, and that other sequence-specific protectionmechanisms (like antisense RNA or silencing) are unlikely to be successful.ReferencesAzzam O, Yambao Ma. LM, Muhsin M, McNally K, Umadhay K. 1999. Genetic diversity ofrice tungro spherical virus in tungro-endemic provinces of the Philippines and Indonesia.Arch. Virol. (In press)Cabauatan PQ, Melcher U, Ishikawa K, Omura T, Hibino H. Koganezawa H, Azzam O. 1999.Sequence changes in six variants of rice tungro bacilliform virus and their phylogeneticrelationships. J. Gen. Virol. 80(8):2229–2237.Villegas LC, Druka A, Bajet NB, Hull R. 1996. Genetic variation of rice tungro bacilliformvirus in the Philippines. Virus Genes 15:l–7.NotesAuthors’addresses: O. Azzam. F. Sta. Cruz, E.L. Coloquio. International Rice Research Institute,MCPO Box 3127, Makati City 127l, Philippines: A. Kliiti, J. Futterer, I. Potrykus,Institute of Plant Sciences, ETH, CH-8092 Zurich. Switzerland; and R. Hull, John InnesCentre, Colney, Norwich NR4 7UH, UK.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.44 Azzam et al

Multilocation evaluation of advancedbreeding lines for resistanceto rice tungro virusesR.C. Cabunagan, E.R. Angeles, S. Villareal, O. Azzam, P.S. Teng, G.S. Khush, T.C.B. Chancellor,E.R. Tiongco, X.H. Truong, S. Mancao, I.G.N. Astika, A. Muis, A.K. Chowdhury, V. Narasimhan,T. Ganapathy, and N. SubramanianTwelve advanced breeding lines with different sources of resistance againstrice tungro viruses were tested together with checks IR62 and IR64 inreplicated 8 x 8-m plots at six locations in the Philippines, Indonesia, andIndia from 1995 to 1998. Advanced breeding lines with resistance derivedfrom Utri Merah (IRGC accession no. 16680) had the lowest infection withrice tungro bacilliform virus and rice tungro spherical virus in each of thethree countries, suggesting that the resistance is likely to be effective in awide range of locations. Some of these lines also had promising yieldpotential. Two lines derived from ARC11554 (IRGC accession no. 21473)showed promising results in the Philippines and Indonesia but not in India.The rapid spread of high-yielding rice varieties and the intensification of rice cultivationin South and Southeast Asia since the 1960s resulted in outbreaks of several virusdiseases. Rice tungro is the most destructive of these diseases and can cause largelosses over extensive areas. Breeding for resistance to tungro is an important componentof rice varietal improvement programs in South and Southeast Asia and at IRRI(Khush and Virmani 1985). Until recently, the breeding strategy for tungro resistanceat IRRI was based on using vector resistance. Since 1969, most IR varieties targetedfor the irrigated lowlands have had at least one parent with resistance to the majorgreen leafhopper vector ( Nephotettix virescens ). The main donors have been Ptb18,Gam Pai 30-12-15, and Ptb33. Such varieties escape tungro infection in the fieldunder light to moderate tungro and vector pressure but succumb to infection whenthere are strong sources of inoculum and vectors are abundant (Cabunagan et al 1987).The reaction of some varieties has changed from resistant to susceptible following achange in the virulence of N. virescens in the field (Dahal et al 1990).Currently, the breeding program for tungro resistance at IRRI uses ARC11554(accession no. 21473), Utri Merah (accession no. 16680), Utri Rajapan (accession no.16684), Habiganj DW 8 (accession no. 1175l), and some wild rice as virus-resistantdonors. Genetic studies have also been conducted to investigate the inheritance ofresistance (Imbe et al 1995). New breeding lines have been developed using the mostpromising sources of virus resistance.Preliminary studies were conducted where test entries were inoculated by thetest tube inoculation method in the greenhouse to select promising lines (Angeles etal 1998). Field evaluation of promising lines was conducted in a tungro hot spot inMidsayap, North Cotabato, Philippines. Some of the most promising lines from testcrosses in these trials were selected for further evaluation in replicated field trials inareas with reported high tungro disease incidence in the Philippines, Indonesia, andIndia. Field trials were conducted in collaboration with PhilRice, the Agency for AgriculturalResearch and Development (AARD) in Indonesia, and the Indian Council

of Agricultural Research (ICAR). Preliminary results from early trials were reportedin Cabunagan et al (1996, 1998). Trial data from India and Indonesia are presentedelsewhere in this volume (Astika, Chowdhury, Subramanian et al). In this paper, wesummarize results from trials in the Philippines and attempt to provide a synthesis ofthe findings from the different locations.Experimental studiesTest locationsTrials in the Philippines were carried out on experimental farms of PhilRice inMaligaya, Nueva Ecija Province, and in Midsayap. North Cotabato Province. In Indonesia,trials were conducted on the experimental farms of the Food Crop ProtectionCenter VI Field Laboratory in Celuk, Gianyar District, Bali, and of the Mares ResearchInstitute for Maize and Other Cereals in Maros and in Sidrap District, SouthSulawesi. In India, trials were conducted on the Regional Rice Experimental Farm ofBidhan Chandra Krishi Viswavidyalaya in Chakdah, West Bengal, and on the experimentalfarm of Tamil Nadu Agricultural University Rice Research Station in Tirur,MGR District, Tamil Nadu. The trials began in 1995 and continued until 1998 (Table1). At least three trials were conducted at each site. with the exception of Maros (twotrials) and Sidrap (one trial). In the Philippines, trials were conducted in both the wetand dry seasons. In India and Indonesia, most of the trials were carried out in the wetseason, when tungro incidence was greater. Four sets of trials were conducted, eachcovering at least three sites, with some lines and varieties evaluated in two or moresets of trials (Table 1, Figs. 1-4).Test lines and varietiesAdvanced breeding lines evaluated in the trials were developed in the IRRI breedingprogram. Table 2 lists these lines and their parents. Two of the virus-resistant parentsused in the crosses, Balimau Putih and Utri Merah, are susceptible to N. virescens ingreenhouse tests conducted at IRRI using vector populations collected in Los Baños.The other two virus-resistant parents, ARC11554 and Oryza longistaminata, are resistantto N. virescens. Varieties IR62 and IR64 were used as field-resistant and susceptiblechecks, respectively. Both varieties were resistant to N. virescens when firstreleased, but in many areas IR64 is now susceptible to N. virescens and succumbs totungro disease where inoculum sources are present.Experimental design and data collectionIn each trial, a randomized complete block design was used with four replications.The plot size was 8 × 8 m with a 2-m distance between plots. Two to three seedlingsper hill were transplanted at 21 d after sowing at 20 × 20-cm spacing and exposed tonatural infection with tungro viruses. In Celuk, Maros, Chakdah, and Tirur, spreader46 Cabunagan et al

Table 1. Infection with rice tungro viruses, tungro disease incidence, and green leafhopper(GLH) numbers on susceptible check IR64 at 60 days after transplanting in multilocation fieldtrials in India, Indonesia, and the Philippines, 1995-98.Set Location Season/year1 Maligaya, PhilippinesMaligaya, PhilippinesMidsayap, PhilippinesMidsayap, PhilippinesMidsayap, PhilippinesBali, IndonesiaWS 95WS 96WS 95DS 96WS 96DS 9569.325.099.363.880.076.00.37.30.020.02.31.5Bali, Indonesia41.0Maros, IndonesiaMaros, IndonesiaTamil Nadu, IndiaTamil Nadu, IndiaWest Bengal, IndiaMaligaya, PhilippinesMaligaya, PhilippinesMidsayap, PhilippinesMidsayap, PhilippinesBali, IndonesiaTamil Nadu, IndiaWest Bengal, IndiaMaligaya, PhilippinesMidsayap, PhilippinesBali, IndonesiaMaligaya, PhilippinesMidsayap, PhilippinesSidrap, IndonesiaTamil Nadu, IndiaWest Bengal, IndiaWS 96DS 95WS 96DS 96WS 96WS 96DS 97WS 97DS 97WS 97WS 97WS 97WS 97Infection a (%)BS B S VIS1.010.326.50.34.83.026.553.854.567.821.327.00.83.5DS 98 0.5 0.0DS 98 84.3 3.5WS 98 59.8 4.0WS 98 0.0 1.5WS 98 78.0 6.0WS 98 44.5 1.5WS 98 0.0 0.5WS 98 0.0 0.030.59.00.88.314.518.529.05.019.539.55.810.56.860.314.533.320.020.337.53.58.828.02.312.844.32.51.562.526.5100.099.3100.099.860.30.56.0 11.02.337.30.317.01.532.58.33.51.346.315.091.03.8100.02.099.85.070.55.564.30.599.387.50.097.3100.00.0ndGLH(no.)7.078.856.5179.5117.52.08.811.1318.3nd9.39.848.828.849.838.560.312.835.860.085.030.317.8172.526.04.0nda BS = RTBV + RTSV, B = RTBV alone, S = RTSV alone, VIS = visual assessment of disease, nd = no data. WS = wetseason, DS = dry season.Fig. 1. Mean infection with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus(RTSV), tungro disease incidence and numbers of green leafhopper (GLH) vectors in the firstset of multi-location field trials.Multilocation evaluation 47

Fig. 2. Mean infection with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus (RTSV),tungro disease incidence, and numbers of green leafhopper (GLH) vectors in the second set ofmultilocational field trials.Fig. 3. Mean infection with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus(RTSV), tungro disease incidence, and numbers of green leafhopper (GLH) vectors in the third set ofmultilocational field trials.rows of an N. virescens- and tungro-susceptible variety were placed between the fourblocks to enhance disease spread. No insecticide was applied to the seedbed or fieldplots during the trials. Fertilization and other management practices were based onrecommendations in the respective test locations.Plants were assessed for disease symptoms and leaves sampled for detection oftungro viruses by enzyme-linked immunosorbent assay (ELISA) at 30-35 and 55-60d after transplanting. Disease was recorded and leaves were sampled in six quadratsof 4 × 4-m hills in a "W" pattern in each plot. Disease assessment was based onsymptoms of stunting and yellowing. Six- to 8-cm-long leaf samples were taken fromrice plants and placed individually in plastic sleeves for temporary storage. Leafsamples were sent by courier to IRRI where the ELISA test was carried out in thevirology laboratory. Leafhopper vectors were collected using 10 sweeps of a 30-cmdiameterinsect net in each plot on the same dates as for disease assessment.48 Cabunagan et al

Fig. 4. Mean infection with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus (RTSV),tungro disease incidence, and numbers of green leafhopper (GLH) vectors in the fourth set ofmultilocational field trials.Table 2. Advanced breeding lines evaluated for resistance to rice tungro viruses in multilocationfield trials in India, Indonesia, and Philippines, 1995-98.Breeding linelR68305-18-1lR69705-1-1-3-2-1lR69726-16-3-2lR69726-116-1-3lR69734-5-1-2lR69734-128-2-3lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1CrossIR64 *4/Balimau Putih alR1561-228-3-3 *2/Utri MerahlR61009-37-2-1-2///IR1561-228-3-3/Utri Merah//lR1561lR61009-37-2-1-2///lR1561-228-3-3/Utri Merah//lR1561lR44624-127-1-2-2-3///IR1561-228-3-3/Utri Merah//lR1561lR44624-127-1-2-2-3///lR1561-228-3-3/Utri Merah//lR1561lR1561-228-3-3 *2/ Oryza longistaminatalR1561-228-3-3 *6/ARC11554lR1561-228-3-3 *6/ARC11554lR1561-228-3-3 *3/Habiganj DW8//4* IR64lR1561-228-3-3 *2/Utri Merah//lR24IR64/ O. rufipogon //3* IR64a Varieties underlined are the virus-resistant donors.In selected trials, yield data were obtained from a 3 × 3-m sample area at harvest.Yields were adjusted to 14% moisture content.Data analysisDifferences between test lines and varieties in tungro virus infection, tungro diseaseincidence (percentage of diseased hills), and GLH numbers were analyzed using analysisof variance (ANOVA). For each variable analyzed, values represented the averageof two sampling dates. Infection with tungro viruses [rice tungro bacilliform virusMultilocation evaluation 49

(RTBV) and rice tungro spherical virus (RTSV)] was treated as total RTBV (bothRTBV + RTSV and RTBV alone) and total RTSV (both RTBV + RTSV and RTSValone). Means were separated using Duncan’s multiple range test.ResultsVariation in tungro incidence and GLH abundanceConsiderable variation occurred in tungro incidence and GLH numbers between testlocations, seasons, and years. This is illustrated by data for tungro incidence andGLH abundance on the susceptible check IR64 (Table 1). In some trials (highlightedin Table 1), disease incidence was too low to allow test entries to be evaluated effectively.In Midsayap and in Bali, tungro incidence was consistently high in both the wetand dry seasons, ranging from 91% to 100%. In Maligaya, tungro incidence wasgreater in the wet season (WS) than in the dry seasons (DS). At other sites, wheremost of the trials were conducted only in the wet season, tungro incidence was generallylow, except for the 1997 trials in Tamil Nadu and West Bengal and the 1998 trialin Sidrap. Green leafhoppers were often most abundant at each of the two sites in thePhilippines, but there was no apparent relationship between leafhopper numbers andtungro incidence and no clear seasonal trend in abundance. For example, tungro incidencein Maligaya was low in both the 1997 and 1998 DS when GLH numbers wererelatively high. In contrast, GLH numbers were low in Bali during the 1995 DS and1996 WS, but disease incidence in both seasons was high.Reaction of test lines in Maligaya and MidsayapTable 3 shows the results from the first set of test entries evaluated in three trials inMidsayap and two trials in Maligaya. The resistant check, IR62, had a consistentlylow infection with RTBV, the highest incidence of tungro in IR62 was 21% in Midsayapin the 1996 WS. Although infection with RTSV was quite high in some trials, theperformance of IR62, which has been grown for many years in Midsayap District,showed that leafhopper resistance can play an important role in reducing tungro incidence.In subsequent trials (Tables 4 and 5), tungro incidence in IR62 was also lowwith the exception of the 1997 WS trial in Midsayap, when it reached 47% underconditions of very heavy disease pressure.IR69705-1-1-3-2-1 showed consistently low infection with RTSV and RTBVand low disease incidence (Table 3). Because of its potential, this line was included inall subsequent trials with similar results (Tables 4 and 5). Its performance showedthat resistance to RTSV and resistance to multiplication of RTBV had been successfullytransferred from its parent Utri Merah. IR68305-18-1 had a high RTSV infectionand moderate RTBV infection (Table 3). Symptoms on this line, however, werenot severe and plants exhibited some tolerance, like resistant parent Balimau Putih.IR71030-2-3-2-1 had low infection with tungro viruses (Table 3). Tungro incidencein this line was generally low, although it reached 30% in the 1996 DS in50 Cabunagan et al

Table 3. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus(RTSV), visual disease incidence, and green leafhopper (GLH) vector numbers on rice varieties andadvanced breeding lines tested in Maligaya and Midsayap, Philippines, 1995 wet season (WS) and1996 WS and dry season (DS).Variable Variety/line Maligaya MidsayapWS 95 WS 96 WS 95 DS 96 WS 96RTBV b infection (%)RTSV infection (%)Visual infection (%)GLH (no.)IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-10.3 b c 1.3 bc 11.0 d36.0 a 17.9 a 92.3 a3.4 b 0.6 c 23.5 c0.1 b 4.4 b 14.0 cd0.3 b 1.5 bc 76.5 b0.1 b 1.1 c 12.0 cd58.8 c 9.5 bc 50.3 c99.1 a 35.4 a 98.1 a93.0 b 17.4 b 91.4 b0.8 e 5.9 cd 11.9 e55.1 c 1.9 d 97.2 ab19.6 d 2.6 cd 23.6 d0.4 b 0.1 b 14.4 c32.1 a 14.1 a 95.6 a0.1 b 0.0 b 12.4 c0.0 b 0.3 b 3.0 d0.1 b 0.1 b 73.5 b0.1 b 0.0 b 15.5 c41.8 c 7.4 c 6.8 b285.0 ab 53.9 a 51.8 ab251.0 b 39.8 a 29.5 ab314.5 a 35.0 ab 18.3 ab257.0 b 12.6 c 63.8 a84.3 c 19.8 bc 7.0 b3.4 c66.1 a8.5 b2.6 c11.9 b1.4 c25.6 c75.3 a48.9 b3.9 d60.1 b4.5 d17.9 c85.1 a29.3 b1.3 d24.3 bc29.6 b4.9 c125.3 a46.6 b40.1 b31.8 b12.1 c12.3 c83.9 a21.9 b2.6 d27.9 b4.9 d70.6 b94.0 a76.5 b2.3 d65.0 b12.0 c21.0 b95.5 a21.9 b1.0 c26.9 b17.6 b18.8 c125.0 a49.9 b47.0 b38.9 b19.3 ca Average of 2 observations at 30-35 and 55-60 d after transplanting.b RTBV infection = total RTBV + RTSV and RTBVinfection alone, RTSV infection = total RTBV + RTSV and RTSV infection alone as assessed by enzyme-linkedimmunsorbent assay. c In a column for each variable, means followed by a common letter are not significantly differentat the 5% level by Duncan's multiple range tests.Midsayap. IR71030-2-3-2-1 is derived from ARCl1554, which has both virus andleafhopper resistance. Because of its resistance characteristics, ARCl 1554 is consideredone of the most useful donors. Another of its progeny, IR71031-4-5-5-1, wasevaluated in subsequent trials (Tables 4 and 5). This line, which has good yield potential,performed consistently well in these trials. IR71026-3-2-4-3-5-2 was heavilyinfected with both RTSV and RTBV and had high tungro disease incidence in the1995 WS trial in Maligaya (Table 3).IR71605-2-1-5-3, derived from Habiganj DW8, was tested in eight field trials in1997-98 (Tables 4 and 5). Results were variable, with moderately high RTBV infectionand high RTSV infection in some seasons. As with IR68305-18-1. this line doesshow, a degree of tolerance for infection. IR73891-2-1-51, with resistance from O.rufipogon, was evaluated in four trials in 1997 (Table 4) and had high levels of tungroincidence in Midsayap. The remaining lines evaluated were all crosses involvingMultilocation evaluation 51

Table 4. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungrospherical virus (RTSV), visual disease incidence, and green leafhopper (GLH) vectornumbers on rice varieties and advanced breeding lines tested in Maligaya and Midsayap,Philippines, 1997 dry season (DS) and wet season (WS).Variable Variety/line Maligaya MidsayapDS 97 WS 97 DS 97 WS 97RTBV b infection (%) IR62 9.0 a c 1.1 c 7.5 cd 31.5 bIR64 7.4 ab 18.4 a 58.5 a 71.5 alR69705-1-1-3-2-1 2.4 ab 5.7 bc 2.5 d 6.5 clR71031-4-5-5-1 1.9 b 5.8 bc 4.3 d 1.9 clR71605-2-1-5-3 3.1 ab 8.2 ab 20.8 b 31.3 blR73890-1-3-1-4-1 4.6 ab 2.7 bc 14.6 bc 37.2 blR73891-2-1-5-1 5.0 ab 3.0 bc 57.1 a 73.7 aRTSV infection (%) IR62 2.3 b 39.5 c 39.4 c 62.2 cIR64 5.4 b 91.3 a 79.0 a 90.9 alR69705-1-1-3-2-1 2.4 b 5.3 e 1.4 e 2.3 dlR71031-4-5-5-11 1.0 c 17.3 d 8.4 d 5.3 dlR71605-2-1-5-3 7.0 a 75.1 b 56.0 b 81.9 ablR73890-1-3-1-4-1 2.9 b 63.1 b 37.8 c 74.3 bcIR73891-2-1-5-1 1.4 c 5.5 e 47.8 bc 65.9 cVisual infection (%) IR62 0.0 a 0.3 c 12.1 cd 46.8 cIR64 0.3 a 23.9 a 53.9 a 100.0 alR69705-1-1-3-2-1 0.0 a 0.0 d 1.1 e 1.4 flR71031-4-5-5-11 0.2 a 0.3 c 8.5 d 1.8 flR71605-2-1-5-3 0.0 a 4.5 b 17.9 c 27.3 elR73890-1-3-1-4-1 0.2 a 0.4 c 12.9 cd 37.3 dlR73891-2-1-5-1 0.0 a 0.0 d 40.1 b 74.0 bGLH (no.) IR62 4.9 c 20.0 bc 16.0 d 2.7 cIR64 27.4 a 54.4 a 68.0 a 57.0 alR69705-1-1-3-2-1 19.0 ab 30.5 b 38.0 bc 26.6 blR71031-4-5-5-11 7.3 bc 13.9 c 24.0 cd 4.5 clR71605-2-1-5-3 26.4 a 32.5 b 38.8 bc 31.6 blR73890-1-3-1-4-1 10.4 bc 26.3 bc 20.0 cd 6.3 clR73891-2-1-5-1 5.4 c 29.3 b 52.8 ab 61.6 aa Average of 2 observations at 30-35 and 55-60 d after transplanting.b RTBV infection = total RTBV+ RTSV and RTBV alone infection, RTSV Infection = total RTBV + RTSV and RTSV alone infection asassessed by enzyme-linked immunosorbent assay. c In a column for each variable, means followed bya common letter are not significantly different at the 5% level by Duncan's multiple range tests.Utri Merah. IR69726-116-1-3 produced excellent results in four trials in 1998 (Table5). IR69734-5-1-2 and IR69734-128-2-3 also performed well in the 1998 WS trials(Table 5).Overall reaction of test lines at all sitesMean values for tungro viruses infection, tungro incidence, and GLH numbersfor each of the four sets of test entries, pooled across all trial sites, are shown inFigures 1–4. In general, the performance of varieties and advanced breeding lines inIndia and Indonesia was similar to that in the Philippines. Regardless of trial location52 Cabunagan et al

Table 5. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus(RTSV), visual disease incidence, and green leafhopper (GLH) vector numbers on rice varieties andadvanced breeding lines tested in Maligaya and Midsayap, Philippines, and in Sidrap, Indonesia,1998 dry and wet seasons.SeasonVariableRTBV b infection (%)RTSV infection (%)Visual Infection (%)GLH (no.]Variety/lineIR62IR64lR69705-1-1-3-2-1lR69726-16-3-2lR69726-116-1-3lR69734-5-1-2lR69734-128-2-3lR71031-4-5-5-1lR71605-2-1-5-3IR62IR64lR69705-1-1-3-2-1R69726-16-3-2lR69726-116-1-3lR69734-5-1-2lR69734-128-2-3lR71031-4-5-5-1lR71605-2-1-5-3IR62IR64lR69705-1-1-3-2-1lR69726-16-3-2lR69726-116-1-3lR69734-5-1-2lR69734-128-2-3lR71031-4-5-5-1lR71605-2-1-5-3IR62IR64lR69705-1-1-3-2-1lR69726-16-3-2lR69726-116-1-3lR69734-5-1-2lR69734-128-2-3lR71031-4-5-5-1lR716052-1-53DryWetPhilippines Philippines IndonesiaMaligaya Midsayap Maligaya Midsayap Sidrap0.9 a c 6.0 d 3.3 ac 18.6 b 2.6 c0.5 a 80.5 a 3.8 a 60.8 a 57.6 a0.8 a 2.1 de 2.7 a 3.4 cd 1.8 cdd–67.5 b – ––0.5 a 0.5 e 0.0 a 0.6 d 0.5 d––0.0 a 19.5 b 9.0 b––1.4 a 7.8 c 4.4 c0.9 a 2.0 de 1.4 a 2.0 d 1.1 cd0.1 a 15.4 c 1.4 a 18.5 b –1.3 b 27.3 c 3.3 a 55.9 b 26.5 b2.3 ab 92.3 a 5.9 a 83.1 a 91.8 a0.0 c 0.9 de 0.0 a 1.8 de 1.1 cd– 90.4 a – ––0.0 c 0.3 e 0.0 a 0.5 e 0.1 d––0.0 a 1.4 de 2.8 c––0.0 a 3.4 d 2.9 c0.0 c 3.4d 0.0 a 12.1 c 1.6 c1.8 b 61.8b 5.4 a 59.8 b –0.5 a 13.9c 0.0 a 23.4 b 4.6 b0.5 a 82.6a 0.0 a 82.5 a 97.9 a0.0 a 1.0d 0.0 a 1.0 de 1.1 bc– 72.9b – ––0.0 a 2.3d 0.0 a 0.3 e 0.3 c––0.0 a 2.0 cd 2.1 bc––0.0 a 1.5 cde 0.3 c1.0 a0.0 a 5.0 c 1.8 bc1.0 a0.0 a 18.8 b –18.1 c11.5 bc 20.6 c 5.5 b43.1 a16.5 abc 141.3 a 27.0 a34.8 ab12.3 bc 65.0 b 24.8 a–27.5 bc––21.0 bc29.5 abc2.8 d11.9 c18.3 d76.0 a33.9 bcd53.8 b34.1 bcd––22.0 cd40.6 bc–10.6 bc19.8 ab17.6 abc9.4 c22.0 a–53.8 b42.8 bc39.4 bc20.0 c43.1 bc–23.0 a28.0 a29.1 a6.1 b–aAverage of 2 observations at 30-35 and 55-60 d after transplanting. b RTBV infection = total RTBV + RTSV and RTBValone infection. RTSV infection = total RTBV + RTSV and RTSV alone infection as assessed by enzyme-linkedimmunosorbent assay c In a column for each variable, means followed by a common letter are not significantly differentat the 5% level by Duncan's multiple range tests. d Not tested.and season, infection with tungro viruses and tungro incidence were low on four UtriMerahprogenies: IR69705-1-1-3-2-1,IR69726-116-1-3, IR69734-5-1-2, and IR69734-128-2-3.Multilocation evaluation 53

GLH numbers were much lower on IR62 and the two ARC11554 progenies,IR71030-2-3-2-1 and IR71031-4-5-5-1, in the Philippines and Indonesia. In India,however, GLH numbers on IR71030-2-3-2-1 and IR71031-1-5-5-1 were comparablewith those on susceptible check IR64 (Subramanian et al, Chowdhury, this volume).Similarly, tungro incidence on these two lines was also relatively high in Tamil Naduand in West Bengal.Yield of test entriesThe highest yielding lines in trials in Bali, Indonesia, were IR68305-18-1, IR71030-2-3-2-1, and IR71031-4-5-5-1, which produced 5 t ha -1 ; however, these data werefrom unreplicated plots. Nevertheless, results from Bali do provide an indication ofthe yield potential of these lines under conditions of high tungro incidence. IR71031-4-5-5-1 generally produced higher yields than other test entries in Midsayap andMaligaya. Data from Midsayap were confounded by the occurrence of feeding damageto rice plants caused by the black bug, Scotinophara coarctata. Yields from thepromising Utri Merah line, IR69705-1-1-3-2-1, were comparable with those of IR64in spite of the large difference in tungro incidence. In the 1998 DS trial in Midsayap,however, another Utri Merah line, IR69734-5-1-2, yielded 3.8 (± 0.3) t ha -1 comparedwith 2.1 (± 0.1) t ha -1 for IR64. IR69726-116-1-3, also derived from Utri Merah, yielded3.2 (± 0.2) t ha -1 in this trial. In Maligaya, low tungro incidence affected comparisonsof yield data between test entries.DiscussionThe potential of tungro to cause severe yield loss and the lack of effective controlmeasures available to rice farmers account for the continuing importance of the disease.Durable resistance to tungro viruses is now regarded as crucial to any long-termsolution to the rice tungro disease problem in South and Southeast Asia. Geographicalvariation in tungro viruses (Dahal et al 1992, Cabauatan et al 1995) and breakdownof resistance because of changes in the virulence of the leafhopper vector (Dahalet al 1990) have been reported. Thus, multilocation testing of germplasm, varieties,and advanced breeding lines with different types of resistance is being undertaken toguide future deployment strategies and ensure durability of resistance.Results from our studies revealed that there are promising advanced breedinglines that showed low infection with tungro viruses across a range of locations.TR69705-1-1-3-2-1, IR69726-116-1-3, and two other promising Utri Merah-derivedlines showed consistently low infection with RTBV and RTSV at all trial locations inthe Philippines, Indonesia, and India. These results suggested that resistance derivedfrom Utri Merah is likely to be effective against tungro disease in a wide range oflocations. Such varieties with tolerance for RTBV and resistance to RTSV are likelyto have a low incidence of tungro and are poor sources of viruses for spread to neighboringfields. Moreover, IR69726-116-1-3 and IR69734-5-1-2 have promising yieldpotential.54 Cabunagan et al

When infected, RTBV-tolerant varieties have mild symptoms, contain a lowamount of RTBV in plant tissues (Cabunagan et al 1993), and show low yield reduction(Hasanuddin and Hibino 1989). In our study, IR68305-18-1, a progeny of BalimauPutih, showed some degree of tolerance for RTBV; however, this line had high RTSVinfection in Midsayap and Bali when tungro incidence was high in the area. In such asituation. IR68305-18-1 could serve as a virus source for neighboring fields. Farmersin Bali liked this line, however, because of its good eating quality, which is comparableto that of IR64. IR64 is widely grown in the area but is highly susceptible totungro. Farmers in Karangasern Regency have already begun to cultivate IR68305-18-1 extensively, although it has not yet been released as a variety (Astika, this volume).Similarly, IR68305-18-1 proved to be popular with farmers in Tamil Nadu andWest Bengal. India.The performance of the leafhopper-resistant check IR62 demonstrates that vectorresistance can remain effective for long periods. IR62 showed good field resistanceat all sites, although RTSV infection was sometimes high in the presence oflarge amounts of inoculum. Combining vector and virus resistance in a variety mayproduce a more durable resistance than using only one type of resistance. In IR71030-2-3-2-1 and IR71031-4-5-5-1, progenies of ARC11554, which is resistant to both theGLH vector and RTSV (Sebastian et al 1996), the resistance was effective in thePhilippines and Indonesia. These lines, however, did not show strong resistance inIndia. Consequently, more work needs to be done to determine whether ARC11554,which originates from India, is a suitable donor for tungro resistance in that country.The other advanced lines—IR71026-3-2-4-3-5-2, a progeny of the wild riceO. longistaminata, IR71605-2-1-5-3 (Habiganj DW 8 line). and IR73891-2-1-5-1 (fromO. rufipogon )—performed well in the preliminary evaluation in the greenhouse atIRRI but showed a very low resistance level in field trials. This may have been due tothe presence of different strains of the viruses or differences in GLH populationsresulting in increased virulence on these varieties.In conclusion, substantial progress has been made in the identification and fieldevaluation of advanced breeding lines that have shown consistently strong resistanceto tungro disease across several sites. There is scope for further improvement in theselines to increase their yield potential and to incorporate resistance to other pests anddiseases. These lines should prove useful to rice breeders in national agricultural researchprograms in Asia for crossing with varieties developed to suit local requirementsfor characteristics such as grain quality.ReferencesAngeles E. Cabunagan RC, Tiongco ER, Azzam O, Teng PS, Khush GS, Chancellor TCB.1998. Advanced breeding lines with resistance to rice tungro viruses. Int. Rice Res. Notes23(1):17–18.Cabauatan PQ, Cabunagan RC, Koganezawa H. 1995. Biological variants of rice tungro virusesin the Philippines. Phytopathology 85:77–81.Multilocation evaluation 55

Cabunagan RC, Hibino H, Sama S, Rizvi SA. 1987. Resistance of rice plants to Nephotettixvirescens in relation to rice tungro-associated viruses. In: Proceedings of the Workshop onRice Tungro Virus. 24–27 September 1986, Maros, South Sulawesi, Indonesia. IndonesiaMinistry of Agriculture. p 66–76.Cabunagan RC, Angeles ER, Tiongco ER, Villareal S. Truong XH, Astika IGN, Muis A,Chowdhury AK, Ganapathy T, Chancellor TCB, Teng PS. Khush GS. 1996. Evaluation ofrice germplasm for resistance to tungro disease. In: Chancellor TCB, Teng PS, Heong KL,editors. Rice tungro disease epidemiology and vector ecology. IRRI Discuss. Pap. Ser.NO. 19. Manila (Philippines): International Rice Research Institute.Cabunagan RC, Florez ZM, Coloquio EC, Koganezawa H. 1993. Virus detection in varietiesresistant/tolerant to tungro. Int. Rice Res. Notes 18(1):32–23.Cabunagan RC, Angeles ER, Tiongco ER, Villareal S, Azzam O. Teng PS, Khush GS, ChancellorTCB, Truong XH, Mancao S, Astika IGN, Muis A. Chowhury AK, Ganapathy T,Subramanian N. 1998. Multilocation evaluation of promising advanced breeding lines forresistance to rice tungro viruses. Int. Rice Res. Notes 23(1): 15–16.Dahal G, Hibino H, Cabunagan RC, Tiongco ER, Florez ZM. Aguiero VM. 1990. Changes incultivar reaction to tungro due to changes in “virulence“ of the leafhopper vector. Phytopathology80:659–665.Dahal G, Dasgupta I, Lee G, Hull R. 1992. Comparative transmission of, and varietal reactionto, three isolates of rice tungro virus disease. Ann. Appl. Biol. 120:287–300.Hasanuddin A, Hibino H. 1989. Grain yield reduction, growth retardation, and virus concentrationin rice plants infected with tungro-associated viruses. Trop. Agric. Res. Ser. 22:56–73.Imbe T, Ikeda R, Kobayashi N, Ebron LA, Yumol RR. Bautista NS, Tambien RE. 1995. Geneticstudies in relation to breeding rice varieties resistant to rice tungro disease. In: Thedevelopment of stabilization technology for double cropping in the tropics. Final report ofthe IRRI-Government of Japan Collaborative Research Project.Khush GS, Virmani S. 1985. Breeding rice for disease resistance. In: Russel GE, editor. Progressin plant breeding. London: Buttenworths. p 239–279.Sebastian LS, Ikeda R, Huang N, Imbe T, Coffman WR, McCouch SR. 1996. Molecular mappingof resistance to rice tungro spherical virus and green leafhopper. Phytopathology86:25–30.NotesAuthors’addresses: R.C. Cabunagan, E.R. Angeles, S. Villareal, O. Azzam, P.S. Teng, and G.S.Khush, International Rice Research Institute (IRRI). MCPO Box 3127, Makati City 1271,Philippines; T.C.B. Chancellor, Natural Resources Institute (NRI), University of Greenwich,Central Avenue, Chatham Maritime, Chatham, Kent ME4 4TB, UK; E.R. Tiongco,X.H. Truong, and S. Mancao, Philippine Rice Research Institute (PhilRice), Maligaya,Muñoz, 3119 Nueva Ecija, Philippines; I.G.N. Astika, Food Crop Protection Center VII,Denpasar, J1. D. I. Panjaitan Renon, P.O. Box 88, Denpasar. Bali, Indonesia; A. Muis,Maros Research Institute for Maize and Other Cereals, JI. Dr. Ratulangi, Maros 90514,South Sulawesi, Indonesia; A.K. Chowdhury, Department of Plant Pathology, BidhanChandra Krishi Viswavidyalaya, Kalyani, West Bengal, India; V. Narasimhan, T. Ganapathy,and N. Subramanian, Department of Plant Pathology, Tamil Nadu Agricultural University,Coimbatore 641003, Tamil Nadu, India.56 Cabunagan et al

Acknowledgments: We gratefully acknowledge the support for these studies of Dr. S.R. Obien.Mr. R. Casco, Ir. F.X. Radjijo, Dr. M. Dahlan, Dr. D. Baco, Dr. A. Hasanuddin, and Dr. S.Mukhopadhay. We also thank Mr. C. Lantican for his assistance in data collection for thePhilippine trials and in the serological testing of leaf samples at IRRI.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Multilocation evaluation 57

Prospects of virus-resistant varieties forcontrolling rice tungro disease in BaliI.G.N. AstikaRice tungro disease has been an important constraint to rice production inBali, Indonesia since 1980. Tungro incidence is favored by staggered plantingdates and the cultivation of susceptible varieties such as IR64 andKrueng Aceh. Since 1995, virus-resistant lines bred at IRRI have beenevaluated in replicated field trials in Celuk Gianyar Province. One line withresistance from Balimau Putih, lR68305-18-1, showed useful tolerance toInfection. It has proved to be popular with farmers in Karangasem Regencybecause of its good eating quality and it is now grown widely in this area.Two lines derived from Utri Merah have shown strong resistance in fieldtrials. None of these lines, however, are commercially available as theyhave not yet been cleared for varietal release.IntroductionBali has become not only a tungro-endemic area but also the core of the Indonesiantungro-endemic region where farmers have suffered from tungro attacks since 1980.During a 10-y r investigation (1987-97). tungro incidence fluctuated each year (Fig.1).Rice tungro disease (RTD) is caused by two viruses: rice tungro bacilliform virus(RTBV) and rice tungro spherical virus (RTSV ), The main vector of this disease is thegreen leafhopper (GLH), Nephotettix virescens. This is the dominant leafhopper specieson rice in tropical areas and it is monophagous on rice (Hibino et al 1978).The adoption of large-scale synchronous rice planting combined with a fallowperiod or secondary break crop ( palawija ) is an effective tungro management strategy.This strategy is difficult to implement in Bali, however, because of water supplylimitation and sociocultural constraints. Synchronous rice planting is usually conductedin small-scale areas or at the Subak level (50–200 ha), but some Subaks mayhave staggered plantings. Indeed, Bali Province is still regarded as an asynchronouslyplanted region. Synchronous planting enables GLH numbers to increase continuouslyin each paddy plot until harvesting time and then decline. In contrast. in asynchronousplanting areas. after the first (G1) generation, the population density of GLHdecreases sharply because of dispersal activity (emigration). GLH migrate to youngerrice plants and transmit RTD from older diseased rice plants. This mechanism facilitatesthe spread of RTD in asynchronous areas (Fig. 2. Aryawan et al 1993).Some control strategies have been implemented in Bali, such as a planting regulationenforcing sowing 5 d after first land preparation/plowing, selective eradication,and pesticide use. These measures, however, have not given satisfactory results andRTD remains a big problem in rice production in Bali. Therefore, using RTD-resistantvarieties with high yield and good quality is a promising alternative for tungromanagement in Bali.

Fig. 1. Seasonal and annual fluctuations in the area affected by rice tungro disease in Bali province,Indonesia, 1988-98.Fig. 2. Population development of green leafhopper (GLH) adults on Krueng Aceh (dotted lines) andIR36 (solid lines) in asynchronous and small-scale synchronous rice cropping areas during the 1987dry season and the 198788 wet season (Aryawan et al 1993).60 Astika

Tungro in Bali (1987-97)Tungro infestation in Bali during the past 10 years has fluctuated from season toseason. Tungro incidence is commonly higher in the wet season than in the dry season(Fig. 1). The peak tungro incidence in the dry and wet seasons occurs in May andJanuary, respectively (Fig. 3).Although all of the regencies in Bali have been attacked by RTD, the endemicregencies are Badung, Tabanan, and Gianyar, where RTD incidence exceeded 200 hayear -1 more than five times during 1987-97 (Fig. 4).Rice varieties used in BaliRice varieties grown and rainfall strongly influence RTD incidence. Rainfall is difficultto forecast, however, and recently it has been difficult to make a clear distinctionbetween dry and wet seasons.In 1990-91, which was the peak of the RTD problem between 1987 and 1997. thedominant rice variety was IR64. As the performance of IR65 has been disappointingin Bali. the Indonesian Government has recommended growing resistant varietiessuch as IR66 and IR72. In 1992-93, RTD incidence increased again and this coincidedwith the dominance (70%) of IR64 in the field (Fig. 5).Fig. 3. Monthly populations of green leafhoppers (GLH) caught in light traps and monthly rainfall andtungro incidence in Bali Province, Indonesia, 1987-97.Prospects of virus-resistant varieties 61

Fig. 4. Frequency of occurrence of tungro in affected areas (exceeding 200 ha yr -1 ) in Bali Province,Indonesia, 1987-97.The three regencies of Tabanan, Badung, and Gianyar have been endemic RTDareas since the 1997-98 wet season and the 1998 dry season. An increase in the cultivationof susceptible IR64 was followed by an increase in the incidence of RTD in thefield. Although IR64 is very susceptible to tungro in Bali, farmers prefer it because ofits high production potential, good taste, and high market price. Farmers still do nothave market access to resistant varieties that are high-yielding and have good taste.Moreover, experience has shown that resistant varieties eventually succumb to tungroafter being used continuously for several cropping seasons (Table 1). For example, inthe 1992-93 cropping seasons, IR66 was moderately resistant to RTD but in 1995-96it became susceptible. IRRI has produced mainly GLH-resistant varieties, but theyare likely to become susceptible after continuous use for several cropping seasons(Inoue and Ruay-Aree 1977). It will be more effective if virus-resistant rice varietiesare developed (Dahal et al 1990).From 1995 until 1998, IRRI, the Natural Resources Institute (UK), and the FoodCrop Protection and Horticulture VII in Denpasar collaborated on tungro diseasemanagement. In the Celuk field laboratory, rice lines and varieties were evaluated forresistance to tungro viruses (Tables 2 and 3). IR68305-18-1 showed susceptibility toRTBV in the 1995 dry season trial, but tungro disease incidence on the variety wasrelatively low, indicating that the line has some tolerance for infection. One line derivedfrom Utri Merah, IR69705-1-1-3-2, showed consistently strong resistance totungro viruses in all trials. Another Utri Merah line, IR69726-116-1-3, performedwell in the latest trial in the 1998 wet season and has an excellent plant type.Yields of these lines ranged from 4.6 to 5.5 t ha -1 . In a taste test, 90% of respondentssaid that the taste of three of the new lines (IR69726-116-1-3, IR69705-1-1-3-2-1, and IR68305-18-1) was excellent and similar to that of IR64. IR68305-18-1 is62 Astika

Fig. 5. Proportion of varieties grown in Tabanan, Badung, and Gianyar regencies during 1987-97 in Bali, dry and wet seasons.Prospects of virus-resistant varieties 63

Table 1. Reaction of rice varieties to tungro disease based on IRRI standard evaluations ain five cropping seasons at Badung and Gianyar regency.VarietyCisadaneCikapundungIR26Bengawan SoloIR36IR42IR64CiliwungIR66IR72Krueng AcehMarosSadangMembramoIR62LariangIR74CendranaeBarumunIR68Categories b1992-93 1993-94 1994 1995-96 1997-98SMR–SS–––S––––S––––S––––S––––S–SSSSS–––S–MRMRMSSSRR–S––MS––S––––S––––S––––S––––SS–MS––RRR–RS–S–––RR––––S––a Chaudhary (1996).b R = resistant, MR = moderately resistant, MS = moderately susceptible, S =susceptible.Table 2. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungro sphericalvirus (RTSV) on rice varieties and advanced breeding lines at Celuk, Bali, Indonesia, in the1995 dry season (DS) and 1996-98 wet seasons (WS).VariableRTBV bRTSVVariety/LineIR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69726-16-3-2lR69726-116-1-3IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69726-16-3-2lR69726-116-1-3DS 953.6 d c72.9 a34.8 c5.4 d55.0 b3.1 d– d–––––12.5 c90.3 a71.8 b2.5 d82.8 a5.6 cd––––––WS 96 WS 97 WS 983.9b23.0 a1.5 bc1.9 bc4.4 b0.5 c––––––1.5 c69.8 a–2.0 c––1.3 c10.0 bc24.0 b24.3 b––4.1 c65.3 a–0.4 d––0.1 d5.5 c––29.9 b0.6 cd5.6 cd39.3 a11.0 c1.4 d21.4 b3.1 d––––––8.8 c87.8 a–3.5 c––5.5 c55.0 b36.0 b34.3 b––14.6 c84.5 a–0.0 d––0.6 d46.0 b––58.1 b0.5 da Average of 2 observations at 30-35 and 60-65 days after transplanting.b Values for RTBV and RTSV are the combinedtotals of single and double infections. c In a column for each variable, means followed by a common letter arenot significantly different at the 5% level by Duncan's multiple range test. d Indicates variety or line not tested.64 Astika

Table 3. Percent incidence of rice tungro disease (RTD) and numbers of green leafhopper(GLH) vectors a on rice varieties and advanced breeding lines at Celuk, Bali, Indonesia,in the 1995 dry season (DS) and 1996-98 wet seasons (WS).Variable Variety/line DS 95 WS 96 WS 97 WS 98Visual (%)GLH no. bIR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-23.3 d c96.0 a29.6 c1.9 d57.6 blR71030-2-3-2-11.0 dlR71031-4-5-5-1– dlR71605-2-1-5-3 –lR73890-1-3-1-4-1 –lR73891-2-1-5-1 –lR69726-16-3-2 –lR69726-116-1-3 –IR621.0 aIR644.3 alR68305-18-11.8 alR69705-1-1-3-2-12.5 alR71026-3-2-4-3-5-2 3.5 aIR71030-2-3-2-10.5 alR71031-4-5-5-1 –lR71605-2-1-5-3 –lR73890-1-3-1-4-1 –lR73891-2-1-5-1 –lR69726-16-3-2 –lR69726-116-1-3 –2.5 bc32.3 a1.0 a0.4 c4.9 b0.4 c––––––3.0 a6.9 a1.0 a2.8 a6.8 a2.6 a––––––2.3 d99.8 a–0.8 d––1.3 d10.3 cd18.5 bc35.3 b––2.8 c60.3 c–11.3 bc––1.3 c18.5 bc21.0 bc39.5 ab––6.6 c83.9 a–0.9 c––0.9 c6.8 c––34.5 b0.5 c3.3 c18.8 a–15.6 ab––2.4 c11.4 b––18.0 a15.0 aba Average of 2 observations at 30-35 and 60-65 days after transplanting.b Numbers of GLH per 10sweeps of a 30-cm-diameter insect net. c In a column for each variable. means followed by a commonletter are not significantly different at the 5% level by Duncan's multiple range test. d Indicates varietyor line not tested.tolerant of RTBV, whereas IR69726-116-1-3 and IR69705-1-1-3-2-1 are resistant toRTSV and tolerant of RTBV. IR68305-18-1 is now grown widely in KarangasemRegency and has shown good field resistance. The three lines have not yet been released,however, and so they cannot be recommended officially to farmers.ConclusionsResults indicated that these three lines have good potential for controlling tungrodisease in Bali: IR68305-18-1 in the dry season and IR69726-116-1-3 and IR69705-1-1-3-2- 1 in the wet season.Prospects of virus-resistant varieties 65

ReferencesAryawan GN, Widiarta N, Suzuki Y, Nakasuji F. 1993. Life table analysis of the green riceleafhopper, Nephotettix virescens (Distant) (Homoptera: Euscelidae), in small-scale synchronousand asynchronous rice fields. Appl. Entomol. Zool. 28:390–393.Chaudhary RC. 1996. Standard evaluation system for rice. Genetic Resources Center, InternationalRice Research Institute, Los Baños, Laguna.Dahal G, Hibino H, Cabunagan R, Tiongco ER, Flores ZM. Aguiero VM. 1990. Changes incultivar reaction to tungro due to changes in 'virulence' of the leafhopper vector. Phytopathology80:665–695.Hibino H, Roechan N, Sudarisman S. 1978. Association of two types of virus particles withpenyakit hambang (tungro disease) of rice in Indonesia. Phytopathology 68:1412–1416.Inoue H. Ruay-Aree S. 1977. Bionomics of green rice leafhopper and epidemics of yelloworange leaf virus disease in Thailand. In: Tropical Agricultural Research Series No. 10.Japan: Tropical Agriculture Research Center (TARC). p 117–121.NotesAuthor’s address: I.G.N. Astika, Food Crop Protection Center VII, Denpasar, JI. D. I. PanjaitanRenon. P.O. Box 88, Denpasar, Bali. Indonesia.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998. IRRI, Los Baños, Philippines. Makati City Philippines): InternationalRice Research Institute. 166 p.66 Astika

Evaluating rice germplasm for resistanceto rice tungro disease in West Bengal,IndiaA.K. ChowdhuryRice tungro disease is a recurrent problem in areas of West Bengal, Indiawhere rice is grown continuously. Rice germplasm developed at IRRI withresistance to tungro viruses were evaluated from 1995 to 1998 at theRegional Rice Research Station of Bidhan Chandra Krishi Viswavidyalayaat Chakdah. Line lR69705-1-1-3-2, with resistance derived from Utri Merah,had consistently low infection with rice tungro spherical virus and rice tungrobacilliform virus in three trials. lR68305-18-1 has a good plant type andexcellent eating quality and is also considered to be a promising line. Thedeployment of tungro-resistant varieties is considered to be the most importantstrategy for managing tungro disease in the future.Rice tungro is one of the most common diseases occurring in both aman (winter) andwet-season rice. Rice is the principal crop in the wet season in West Bengal. India,where more than 90% of cultivated land is planted with high-yielding or tall indicavarieties. Tungro disease incidence is highest in the aman season, which coincideswith the peak abundance of vectors, mostly rice green leafhopper (GLH). The diseaseoccurs sporadically, but it can become widespread. Under field conditions. the mostconspicuous tungro symptoms are usually found only in a few high-yielding and localvarieties and can easily be confirmed. In other cases, although infection by tungroassociatedviruses occurs, plants do not develop any characteristic symptoms, whichcauses problems in diagnosis by visual observation. In many areas of West Bengal,rice is grown continuously depending on the availability of irrigation water. In theseareas, tungro is a recurrent problem, unlike in rainfed single-cropped areas, even thoughvery low populations of GLH occur during the winter and summer months.In collaboration with IRRI and the Natural Resources Institute (UK). ricegermplasm accessions have been tested for resistance to rice tungro viruses since the199.5 wet season. This work has been conducted at the Regional Research Station ofBidhan Chandra Krishi Viswavidyalaya, in a new alluvial zone at Chakdah in WestBengal, India. The methodology of the experiments followed the protocol describedby Cabunagan et al (this volume). To coincide with the natural incidence of GLH,transplanting was done in late July. No plant protection chemicals were applied. Indexingof leaf samples by enzyme-linked immunosorbent assay (ELISA) was done inthe plant virology laboratory at IRRI.In the 1995 trial, five rice varieties and one advanced breeding line were evaluatedfor resistance against tungro disease (Table 1), The susceptible check was IR64and the field-resistant check was IR62. Two unimproved resistant donors. BalimauPutih and Utri Merah, and IR26, which has resistance to rice tungro spherical virus(RTSV), were included in the trial. In addition, IR68305-18-1, a cross between IR64and Balimau Putih, was evaluated. Because infection with tungro viruses was low in

Table 1. Percent infection with rice tungro bacilliform virus (RTBV) and rice tungro sphericalvirus (RTSV), visual tungro disease incidence, and number of green leafhoppers (GLH) a onrice varieties and advanced breeding lines at Chakdah, West Bengal, India, 1995 wet season.VariableBB c 9.3 a 4.3 abSS 13.4 a 3.3 bVisual 7.0 b 7.8 bGLH d (no. 10 85.3 a 20.6 bsweeps -1 )Varieties/lines bIR26 IR62 IR64 lR68305-18-1 Balimau Putih Utri Merah2.1 b17.0 a31.1 a76.4 a1.8 b10.1 a6.3 b52.1 ab4.4 ab16.6 a10.6 b62.7 ab5.6 ab2.3 b4.6 b68.9 aba Average of 2 observations at 30-35 and 60-65 d after transplanting.b In a row for each variable, means followed bya common letter are not significantly different at the 5% level by Duncan’s multiple range test. c values for RTBV andRTSV are the combined totals of single and double infections. BB =total RTBV Infection, SS = total RTSV infections.d Numbers of GLH 10 sweeps-1 of a 30-cm diameter insect net.the trial, it was difficult to assess the performance of the resistant line and varieties.Utri Merah and IR62 had relatively low RTSV infection. In contrast, the level ofRTSV infection in resistant IR26 was similar to that in IR64.In trials between 1996 and 1998, IR62 and IR64 were retained as checks and arange of advanced breeding lines were evaluated for tungro resistance. Infection withtungro viruses was low in both 1996 and 1998, but some differences in the reaction oftest lines were detectable in 1996 (Table 2). A line with virus resistance derived fromvariety ARCl1554, IR71030-2-3-2-1, showed relatively high infection of both RTSVand rice tungro bacilliform virus (RTBV) in the 1996 trial. Similarly, IR71031-4-S-S-1, which also has ARC11554 as a parent, was heavily infected with both viruses in the1997 trial. ARC11554 is of Indian origin and has shown a resistant reaction to tungroviruses in both laboratory and field screening in the Philippines. This variety has beenused in many breeding programs as a tungro-resistant donor. Results from field testingits progeny, however, indicated the need for further evaluation under field andlaboratory conditions using Indian isolates of tungro viruses.IR69705-1-1-3-2, with resistance to RTSV and to multiplication of RTBV fromUtri Merah, had low infection rates for both viruses in all three trials in 1996–98. Itsperformance in 1997, when tungro incidence was high, suggested that it has a strongresistance level. Thus, IR69705-1-1-3-2 is considered to have good potential for growingin West Bengal. IR68305-18-1 is also regarded as a promising line, althoughconditions in the two trials in which it was evaluated were not ideal. This line has agood plant type and excellent eating quality and could be popular with farmers andconsumers.68 Chowdhury

Table 2. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungrospherical virus (RTSV) on rice varieties and advanced breeding lines at Chakdah,West Bengal, India, 1996-98 wet seasons.Variable Variety/line WS 96 b ws 97 WS 98RTBV cRTSVIR62IR64lR68305-18-1lR69705-1-1-3-2-1R71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-3IR62IR64lR68305-18-1lR69705-1-1-3-2-11.5 bc4.3 ab1.1 bc0.3 c0.8 c6.1 a– d–––––7.8 b18.0 a5.1 bc1.1 c2.4 e21.9 b–3.6 de––39.3 a19.1 bc9.4 cd22.0 b––30.1 b50.4 a–1.3 c0 a0 a–0 a––0 a0.3 a––0.1 a0 a0.3 bc1.5 abc–lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-36.0 bc31.1 a––––––––54.0 a64.5 a50.8 a22.0 b––0 c––2.6 ab2.8 a––0 c0 ca Average of 2 observations at 30-35 and 60-65 d after transplanting.b In a column for each variable.means followed by a common letter are not significantly different at the 5% level by Duncan's multiplerange test. c Values for RTBV and RTSV are the combined totals of single and double infections.dindicates variety or line not tested.A comparison of GLH numbers and incidence of tungro viruses and tungro diseasein IR62 and IR64 in the four trials in 1995–98 (Table 3) shows no apparentrelationship between leafhopper numbers and tungro incidence. The data reveal thatGLH-resistant IR62 still has a strong resistance to tungro disease in the field.In conclusion, the deployment of resistant varieties is considered to be the mostimportant strategy in tungro management. The causes of sudden disease outbreakshave yet to be determined. This information would enable forecasting of tungro diseaseand allow the timely application of plant protection measures. Meanwhile, thedeployment of virus-resistant varieties offers good prospects for the future.Evaluating rice germplasm 69

Table 3. Percent incidence of rice tungro disease (RTD) and numbers of green leafhopper(GLH) vectors a on rice varieties and advanced breeding lines at Chakdah, WestBengal, India, 1996-98 wet seasons.Variable Variety/line WS 96 c WS97 WS98RTD (%)incidenceGLH no. bIR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-3IR62IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-39.8 b20.0 a10.8 b9.5 b23.8 a16.9 ab– d–––––26.6 a27.9 a35.9 a24.9 a20.5 a26.0 a––––––12.4 b43.3 a–4.9 b––35.0 a–10.6 b11.8 b––21.1 c40.5 ab–24.3 c––37.3 b38.6 b24.5 c46.5 a––++ e++–++––++++––++++++++–++––++++––++++a Average of 2 observations at 30-35 and 60-65 days after transplanting.b Numbers of GLH 10sweeps -1 of a 30-cm-diameter insect net. c In a column for each variable, means followed by a commonletter are not significantly different at the 5% level by Duncan's multiple range test. d Indicatesvariety or line not tested. e No data.NotesAuthor’s address: A.K. Chowdhury, Bidhan Chandra Krishi Viswavidyalaya, Faculty of Agriculture,Department of Plant Pathology, Mohanpur 741252, West Bengal, India.Acknowledgments: The author gratefully acknowledges financial assistance from the Departmentof International Development through collaboration with NRI and IRRI. Thanksalso go to the Vice Chancellor of Bidhan Chandra Krishi Viswavidyalaya, and researchstaff of RSS Chakdah for assistance.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.70 Chowdhury

Rice tungro disease resistance andmanagement in Tamil Nadu, IndiaN. Subramanian, T. Ganapathy, M. Surendran, and A. Azeez BashaMajor outbreaks of rice tungro disease occurred In Tamil Nadu in 1984and 1992 and the disease has continued to appear sporadically in certaindistricts. Varietal resistance was identified as the most appropriate strategyfor managing tungro and collaborative activities were initiated to evaluatethe performance of advanced breeding lines with resistance or toleranceto tungro viruses. Field trials were conducted at the Rice ResearchStation of Tamil Nadu Agricultural University at Tirur in 1996-98 and severalpromising lines were identified. One tungro-tolerant line, lR68305-18-1, performed well in participatory trials with farmers and is now beingused in a resistance breeding program at Tamil Nadu Agricultural University.Rice tungro disease (RTD) has long been and continues to be a major threat to ricecultivation. It causes great havoc and 30% to 100% losses when it attacks a highlysusceptible variety before flowering in conditions conducive to spread. The disease istransmitted by the green leafhopper (GLH), Nephotettix virescens. Vector control isthe only possible way of managing the disease, but this too is ineffective. Once thedisease has become well established in a field, management by vector control is impossible.Hence, there is a strong case for increased dependence on host-plant resistanceto tungro disease.RTD was first reported in India in 1967 (Raychaudhuri and Ghosh) and outbreakswere reported in Bihar in 1969 and in Kerala in 1973. The disease occurred inTanjore District of Tamil Nadu in 1980 and appeared in Chengalpet District in 1982.Major outbreaks occurred during 1984 and 1992 in Tamil Nadu. In 1984, an epidemicoccurred in Tanjore, Chengalpet, and other districts of Tamil Nadu. At the time, varietieslR50, IR36, and CR1009 showed moderate resistance to the disease in Tanjore.During the 1992 epidemic, RTD was recorded in Thiruvallur, Kanchipuram,Vellore, Villupuram, Thiruvannamalai, and Cuddalore districts. IR50 then showedmoderate susceptibility to RTD, although it showed resistance to GLH. Exploitingresistance in rice cultivars against RTD is a continuing process because the potentialvariability of the causal virus is likely to lead to eventual “breakdown.” Moreover,GLH may adapt to vector-resistant varieties by switching from xylem to phloem feeding,which facilitates both leafhopper population development and virus transmission.Consequently, varietal resistance is not stable and a variety that has high resistanceto GLH may develop a susceptible reaction to RTD.To achieve host-plant resistance to disease, there are various options for usingand deploying resistance genes. Central to any varietal resistance program is the initialfield screening of cultivars under both normal and epidemic conditions. After aresistant line is identified in the field, the material can be used as a donor in a resistancebreeding program.

During the 1984 and 1992 RTD epidemics, researchers at the Rice ResearchStation (RRS) in Tirur identified a few varieties with some resistance to RTD. IR56and BG 367-3 were moderately resistant in 1984, and IR54 and IET 12888 wereresistant in 1992.RTD has continued to occur sporadically in Thiruvallur and Kanchipuram andthere is concern that the presence of inoculum in these areas poses a potential threat toother rice-growing areas, particularly in the Cauvery delta. This led to the developmentof a collaborative research project among Tamil Nadu Agricultural University,the Department of Agriculture, IRRI, and the Natural Resources Institute (UK).From 1996 to 1998, experiments were planted at RRS, Tirur, to assess tungroresistance in rice varieties and advanced breeding lines. The test lines and varietieswere evaluated in small plots arranged in a randomired complete block design (seeCabunagan et al, this volume). Three trials were laid our during the wet seasons (Samba1996-98) and one in the dry season (Navarai 1996). RTD incidence was observed andleaf samples were sent to IRRI to be indexed for the presence of rice tungro sphericalvirus (RTSV) and rice tungro bacilliform virus (RTBV) by enzyme-linkedimmunosorbent assay.On-farm tungro management trials were conducted in 1996-98 in the villages ofVishar and Pudumavilangai to evaluate promising advanced lines under farmers’ fieldconditions. The trials were also used to demonstrate to farmers in village communitiesthe value of tungro resistance as a strategic measure for tungro management. Theperformance of a leafhopper-resistant variety (T 1 ) and a virus-resistant or virus-tolerantline (T 2 ) was compared with that of farmer chosen variety (T 3 ), which was alsoplanted in the farmer’s field. The plot size was 10 x 10 m with a 2-m border row of theleafhopper-resistant variety between plots and surrounding the whole trial area. Eachtreatment had three replicates laid out in a randomized complete block design. Thefarmer was requested not to spray T 1 or T 2 plots against leathoppers or tungro, but hewas free to apply pest management practices of his choice to T 3 plots. Tungro incidencewas assessed by counting the proportion of diseased hills within five 1 × 1 -mquadrats. Leafhopper numbers were estimated by 10 sweeps of a 30-cm-diameterinsect net. Yield data were collected from a 5 × 5-m area in the center of each plot.ResultsField screening for resistance to tungroThe incidence of tungro disease ranged from 0 to 46% on susceptible IR64 in the fourtrials (Tables 1 and 2). No incidence was recorded in the 1998 wet season and incidencewas very low in the 1996 wet season, when there were no significant differencesin virus infection rates among test lines and varieties (Table 1). Data from the1997 wet-season trial could not be analyzed because some treatments had insufficientreplications because of problems with crop establishment.In the 1996 wet season, tungro incidence in ADT37, IR69705-1-1-3-2-1, andIR68305-18-1 was significantly lower than in susceptible check IR64. IR69705-1-1-3-2-1, with resistance to RTSV and RTBV derived from Utri Merah, had low infec-72 Subramanian et al

Table 1. Percent infection a with rice tungro bacilliform virus (RTBV) and rice tungro spherical virus(RTSV) on rice varieties and advanced breeding lines at Tirur, Tamil Nadu, India, in the 1996 dryseason (DS) and 1996-98 wet seasons (WS).Variable Variety/line DS 96 WS 96 WS 97 WS 98RTBV b (%)RTSV (%)IR62/ADT37IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-3IR62/ADT37IR64lR68305-18-1lR69705-1-1-3-2-1lR71026-3-2-4-3-5-2lR71030-2-3-2-1lR71031-4-5-5-1lR71605-2-1-5-3lR73890-1-3-1-4-1lR73891-2-1-5-1lR69734-5-1-2lR69734-128-2-34.4 b c19.4 a2.1 bc0.8 c2.8 bc11.6 a––––––25.8 bc46.0 a27.0 bc16.8 c30.3 b30.5 b––––––0.1 a0.5 a0 a0.3 a0 a0.1 a––––––3.1 a3.8 a0.9 a0.3 a2.9 a1.1 a––––––4.314.8–2.8––3.31.02.82.5––5.826.2–0.7––6.111.85.93.3––0.3 a0.4 a–0 a––0 a0.3 a––0 a0 a0 b1.3 a–0 b––0.3 ab1.1 a––0 b0 ba Average of 2 observations at 30-35 and 60-65 d after transplanting.b Values for RTBV and RTSV are combined totalsof single and double infections. c In a column for each variable, means followed by a common letter are not significantlydifferent at the 5% level by Duncan's multiple range tests. d Indicates variety or line not testedtion with both viruses in the 1996 dry- and wet-season trials when disease levels inthe trials were relatively high. IR68305-18-1, a cross between IR64 and BalimauPutih, also performed well in the 1996 dry-season trial, although infection with RTSVwas relatively high at 27%.IR71030-2-3-2-1, with resistance to tungro viruses and to GLH from ARC11554,did not perform as well as expected. In the 1996 dry-season trial, infection with bothRTSV and RTBV was relatively high. In the 1996 wet season, tungro incidence in thisline was similar to that in IR64. Data from another ARC11554 line evaluated in the1997 and 1998 trials. IR71031-4-55-1, were not conclusive. Results, however, suggestedthe need for further evaluation of the resistance from ARC11553 under fieldconditions in Tamil Nadu.On-farm tungro management trialsIn general, disease incidence in the trials was low. A tungro outbreak occurred inVishar in the 1997 wet season and this provided an opportunity for the GLH- andRice tungro disease resistance 73

Table 2. Percent incidence of rice tungro disease and numbers of green leafhopper (GLH) vectors a onrice varieties and advanced breeding lines at Tirur, Tamil Nadu, India, in the 1996 dry season and1996-98 wet seasons (WS).Variable Variety/line DS 96 WS 96 WS 97 WS 98Visual (%) IR62/ADT37 1.8 bc c 3.1 b 4.8GLH (no.) b IR62/ADT37 – 10.4 a 6.0IR64 23.5 a 10.0 a 36.5lR68305-18-1 0.5 c 3.8 b –lR69705-1-1-3-2-1 0.9 bc 3.6 b 2.0lR71026-3-2-4-3-5-2 1.1 bc 6.8 ab –lR71030-2-3-2-1 5.4 b 9.9 a –dlR71031-4-5-5-1– –4.0lR71605-2-1-5-3 – –2.0lR73890-1-3-1-4-1 – –0.8lR73891-2-1-5-1 – –1.0lR69734-5-1-2 – ––lR69734-128-2-3 – ––IR64 – 11.3 a 7.3lR68305-18-1 – 14.4 a –lR69705-1-1-3-2-1 – 15.1 a 5.9lR71026-3-2-4-3-5-2 – 10.0 a –lR71030-2-3-2-1 – 12.4 a –lR71031-4-5-5-1 – –10.0lR71605-2-1-5-3 – –6.8lR73890-1-3-1-4-1 – –7.4lR73891-2-1-5-1 – –6.5lR69734-5-1-2 – ––lR69734-128-2-3 – ––0 a0 a–0 a––0 a0 a––0 a0 a4.5 a7.1 a–2.5 a––2.4 a2.8 a––1.8 a2.3 aa Average of 2 observations at 30-35 and 60-65 d after transplanting.b Numbers of GLH per 10 sweeps of a 30-cmdiameterinsect net. c In a column for each variable, means followed by a common letter are not significantly differentat the 5% level by Duncan’s multiple range tests. d lndicates variety or line not tested.Table 3. Tungro disease incidence a , green leafhopper (GLH) numbers b , and yield fordifferent lines and varieties in an on-farm trial in Vishar, Tamil Nadu, India, in the 1997wet season.Line/variety GLH (no.) Tungro incidence (%) Yield (t ha -1 )ADT37lR68305-18-1ADT363.8 ± 0.2 17.4 ± 10.2 4.9 ± 0.11.7 ± 0.3 2.6 ± 3.3 6.2 ± 0.46.1 ± 0.6 31.8 ± 13.3 2.5 ± 0.1aNumber of tungro diseased hills at 56 d aftertransplanting (DAT). Mean of three replicatlons. b Averagenumbers of adults and nymphs of Nephotettix virescens per 10 sweeps of a 30-cm-diameter insectnet collected over three sampling dates at 28, 42, and 56 DAT. Mean of three replicatlons.virus-resistant entries to be effectively evaluated. The majority of varieties affected inthe epidemic were the highly susceptible varieties ADT36 and ADT42. ADT36 wasthe variety chosen in the trial by the farmer-collaborator, Mr. Radha Krishnan. Tungroincidence reached 32% by 56 d after transplanting (Table 3). Incidence in the locally74 ubramanIan et al

ecommended vector-resistant variety, ADT37, was 17% at the same date and lessthan 3% in the virus-tolerant IR68305-18-1. This trial stimulated considerable interestamong farmers from both Vishar and other villages who were brought to view thetrials. As a result, farmers from Vishar and Pudumavilangai grew IR68305-18-1 fromseed saved from on-farm trials, which were distributed among them without interventionfrom researchers.ConclusionsA line with resistance to tungro viruses derived from Utri Merah, IR69705-1-1-3-2-1,showed good resistance in on-station field trials. This line has good potential for usein breeding programs in Tamil Nadu. IR68305-18-1 performed well in both on-stationand on-farm trials. This line has a good plant type and grain yield. In addition,farmers like the white and medium-fine grain. IR68305-18-1 is currently being usedin the tungro resistance breeding program at Tamil Nadu Agricultural University.The trials showed that leafhopper resistance can also play an important role intungro management. IR62 and ADT37 both showed good resistance to RTD.ReferenceRaychaudhuri SP, Ghosh A. 1967. Occurrence of paddy virus and virus-like symptoms in India.In: The virus diseases of rice plants. Baltimore, Maryland: Johns Hopkins Press. p 59–65.NotesAuthors’ address: N. Subramanian, T. Ganapathy. M. Surendran, and A. Areez Basha, Departmentof Plant Pathology, Tamil Nadu Agricultural University, Coimbatore 641003, TamilNadu. India.Acknowledgments: This work was done in collaboration with the NRI and IRRI and was fundedby the Department for International Development (UK) through its Crop ProtectionProgramme.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998. IRRI. Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Rice tungro disease resistance 75

Tungro screen kits for extension agentsand plant breedersO. Azzam, L. Kenyon, and P.D. NathFor the last 14 years, IRRI has used the enzyme-linked immunosorbentassay (ELISA) method to screen and evaluate rice germplasm for tungroresistance and tolerance. Unfortunately, due to a lack in resources andtechnical capacities, this technology has not been taken up by nationalprograms. Two years ago, collaborative activities between IRRI and NRIwere initiated to develop diagnostic kits that are simpler and more suitableto the needs of national breeding programs and extension services.In this study, we report on the successful development of a Tungro ScreenKit for rice tungro bacilliform virus, one of the two viruses that cause thetungro disease. Five prototype Tungro Screen B kits were assembled anddistributed at the Tungro Management Workshop held at IRRI in November1998 for field testing in India, Indonesia, and the Philippines.In Southeast Asia, the diagnosis of the two viruses associated with tungro disease,rice tungro bacilliform virus (RTBV) and rice tungro spherical virus (RTSV), reliesmainly on the specific orange-yellow leaf discoloration symptoms exhibited on susceptiblecultivars. Several of these cultivars, however, do not produce the specificsymptoms in the field or under experimental conditions. Furthermore, if plants areinfected with RTSV only, there are no, or only very mild, symptoms.When enzyme-linked immunosorbent assay was introduced at IRRI in 1985, itrevolutionized the approaches to studying the ecology and epidemiology of these twoviruses by providing a relatively fast and accurate means of detection. For the last 14years, IRRI has used the technique to screen and evaluate rice germplasm for tungroresistance and tolerance. More than 35,000 rice accessions have been evaluated andpotential sources of resistance against RTSV and tolerance for RTBV have been identified.In addition to screening germplasm, IRRI assists national research programs intheir studies on tungro disease. Unfortunately, because of a lack of resources for establishingand maintaining such facilities and technical difficulties in producing theantisera locally, and optimizing and troubleshooting the technique routinely, the ELISAprocedure has not been taken up by the national programs. Commercially producedkits based on 96-well ELISA plates and specific antisera against RTBV and RTSV areavailable (e.g., Adgen), but these are generally too expensive for routine use in breedingprograms or by resource-poor extension services. They also usually require someexpertise and the use of expensive equipment for reliable assessment of results.This report summarizes the collaborative activities undertaken by IRRI and theNatural Resources Institute to try to produce antisera with high titer and specificitysuitable for use in simple diagnostic kits by resource-poor national programs andextension services.

Materials and methodsBatches of antisera against the two viruses were produced by using a modification ofthe usual procedure. Instead of giving the rabbits only three intermuscular and oneinterveinal injection of the purified virus particles (in adjuvant) at weekly and biweeklyintervals, they were given nine, with 6-wk intervals between each of the laterseven injections. Approximately 20 mL of blood were collected from each rabbit 10 dafter each of the later seven antigen injections. This immunization regime is reportedto increase the titer and specificity of the resulting antisera (Harlow and Lane 1988).Antisera titers were first measured using purified virus particles in the ring-interfaceprecipitin test (Van Regenmortel 1982). Immunoglobulins were purified fromthe sera by ammonium sulfate precipitation and diethylaminoethyl (DEAE)-cellulosechromatography. Alkaline phosphatase was conjugated to the immunoglobulins bythe glutaraldehyde procedure (Clark and Adams 1977). The titer and specificity ofeach of the different batches of purified immunoglobulins were tested in double antibodysandwich (DAS)-ELISA using IgG to trap and IgG-alkaline phosphatase to detectthe trapped virions.The batches of antiserum with the greatest titer and specificity were tested in anonquantitative membrane-based “tissue-print’’ assay for their suitability for use in asimple diagnostic kit. Freshly cut stems and leaf midribs were “printed” onto thesurface of nitrocellulose or polyvinylidene fluoride (PVDF) membranes. The membraneswere then probed either with the virus-specific antisera batch followed by acommercial alkaline phosphatase-labeled antirabbit immunoglobulin or directly withthe virus-specific antibody batch conjugated to alkaline phosphatase. Signal developmentin either case was by using the 5-bromo-4-chloro-3-indolyl phosphate/nitrobluetetrazolium (BCIP/NBT) substrate.Results and discussionThe need by national programs and extension agents for simple, cheap, and reliablediagnostic tools for rice tungro viruses had been expressed frequently (e.g., Foot 1995).In 1997, Cabauatan and Koganezawa described a simple assay based on the use ofcolored latex beads bound to specific polyclonal antisera. This procedure, however,was not taken up to any great extent, probably because of the requirement to homogenizeand centrifuge samples, and the large quantities of antiserum the test required.Of the batches of antisera produced against RTBV and RTSV by the repeatedimmunization method, RTBV-6 and RTSV-6 showed the greatest titers in the ringinterface test. All seven batches for each virus had sufficient specificity and titer foruse in DAS-ELISA, although they did present relatively high healthy-backgroundabsorbances. This high level of cross-reaction to components from healthy rice plantsmeant that none of the antisera could be used directly in tissue-print assays becausethe difference in signal strength between infected and healthy plants was often toosmall to be reliably observed by the eye.78 Azzam et al

To try to overcome the problem of cross-reaction to healthy plant components,the batches of antiserum with the greatest titer and least cross-reactivity were crossadsorbedwith healthy sap components. The most effective method for doing this wasby first passing the immunoglobulin fractions through a specially prepared healthyrice-proteins-sepharosecolumn (prepared by treating cyanogen-bromide-activatedsepharose with homogenate from healthy rice plants). The resulting solution was thenincubated with a piece of nitrocellulose membrane previously saturated with healthyrice plant homogenate. This mopped up any remaining antibodies with affinity forplant proteins.After this double-cross adsorption, only RTBV-IgG batch 6 retained sufficienttiter to be used effectively with little healthy-background reaction in tissue-print assays(Fig. 1). This doubly cross-adsorbed antibody is the basis for the prototype TungroScreen B diagnostic kits. Because a helper component from RTSV is required forvector transmission of RTBV, a positive reaction with the RTBV detection kit in thefield implies the presence of RTSV as well. A simple diagnostic test for the presenceof RTSV alone, however, would still be useful.Five prototype Tungro Screen B kits were distributed at the November 1998Tungro Management Workshop held at IRRI for field testing in India, Indonesia, andthe Philippines. Included with each kit was a question sheet to be filled out and returnedto IRRI with the test membranes. The questionnaire asked for details aboutrice varieties tested, plant age, how plants were stored prior to testing. if there was adelay between printing and developing the membranes, and how easy the evaluatorfound the test procedure, and comments about the kit or its possible improvement.Based on these responses, it is anticipated that modifications or improvements to thekit will be made in the near future.Fig. 1. Tissue printing results of a series of purified rice tungro bacilliform virus (RTBV) dilutions(Pv), RTBV-infected plant prints (B), rice tungro spherical virus (RTSV)-infected plant prints (S), andvirus ELISA-negative plant prints (N) when hybridized with RTBV IgG without and with cross- adsorption.IgG and conjugate dilutions used were 10 -3 .Tungro screen kits 79

ReferencesClark MF, Adams AN. 1977. Characteristics of the microplate method of enzyme-linkedimmunosorbent assay for the detection of plant viruses. J. Gen. Virol. 34:475–483.Cabauatan P, Koganezawa H. 1997. Alternative methods for detection of rice tungro viruses.In: Chancellor TCB, Thresh JM, editors. Epidemiology and management of rice tungrodisease. Chatham (UK): Natural Resources Institute.Foot C. 1995. A diagnostic kit for the rapid detection of tungro in the field. NRI-IRRI internalreport.Harlow E, Lane D. 1988. Antibodies, a laboratory manual. Sew York (USA): Cold SpringHarbor Laboratory Press.Van Regenmortel MHV. 1982. Serology and immunochemistry of plant viruses. New York(USA): Academic Press. 302 p.NotesAuthors’ addresses: O. Azzam and P.D. Nath, International Rice Research Institute, MCPOBox 3127, Makati City 1271, Philippines; L. Kenyon. Natural Resources Institute,University of Greenwich, Central Avenue, Chatham, Maritime, Kent ME4 4TB, UK.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.80 Azzam et al

Are tungro disease counts repeatable?K.G. SchoenlyScientists often assume that the measurements they take are repeatableacross observers and sites; however, experience from different and unrelateddisciplines indicate that interobserver repeatability varies with thesophistication of the measurement, observer experience, and the measurementscale used. Recent developments in repeatability methodologyfrom quantitative genetics, for example, has produced an easy-to-interpretrepeatability index R that varies from 0 (no repeatability) to 1 (perfectrepeatabilty) derived from a one-way ANOVA (or intraclass correlation) thatmay be useful for plant protection workers. For measurements that arerepeatable across a range of observers, the upper confidence interval forR (e.g., 95%) should equal or approach 1. For tungro disease counts, onestudy from Thailand revealed an R value of 0.1407 and an upper (95%)confidence interval of 0.5501. Although this value is low, without morefield counts from more sites, it is premature to ask if tungro disease countsare repeatable. Although no scientific measurement is expected to haveperfect repeatability, this Thai study underscores the need for plant protectionworkers to conduct frequent repeatability trials of their scientificmeasurements as a routine quality assurance procedure, particularly whenmultiple persons are required at multiple sites and when data from suchstudies are pooled for later statistical analysis.Measurements that scientists take are often assumed to be repeatable and highly preciseacross a range of observers (Krebs 1989); however, interobserver repeatabilityvaries with the sophistication of the measurement, observer experience, and the measurementscale used. Published trials of repeatability for plant injuries in rice arescarce but revealing. In showing possible causes of varietal reaction to rice tungrodisease, Ling (1979) reported that, of 561 rice varieties scored for their reaction totungro by four observers in 1976, only 49% of within-observer readings (based ontwo readings of the same varieties) scored identically. Interscorer results from 28 ricevarieties, taken by six scorers in 1978, showed that 43% of the varieties differed by 2points on a 5-point scoring scale. This study also showed that scoring ability can beimproved by experience. For field surveys, Ling (1979) recommended that data recordingfor tungro be restricted to one person using a tape recorder and suggested thata data transfer method be used instead of employing multiple persons to record dataat the same site. Studies that require the same data to be collected at multiple sites,however, will likely employ multiple persons. If data from such studies are pooled forstatistical analysis, interobserver repeatability becomes an unavoidable scientific issue.Recent developments in repeatability methodology from quantitative genetics(Becker 1984, Lessells and Boag 1987) have produced an easy-to-interpret repeatabilityindex R (based on a one-way ANOVA classification; see Krebs 1989 for workingexamples) that varies from 0 (no repeatability) to 1 (perfect repeatability). R isalso known as the intraclass correlation, accessible in standard biometrics textbooks

(e.g., Sokal and Rohlf 1995, p 213; Zar 1984, p 323-325). and is calculated as R = S 2 A /(S 2 E + S 2 A ), where S 2 A is the variance among items and S 2 E is the variance withinindividuals. Thus, if measurements are perfectly repeatable. S 2 E is zero and R = 1.0.When R is computed with its 95% confidence limits, for example, one hopes that theupper confidence limit of R equals or closely approaches 1. Few scientific measurementsare expected to have perfect repeatability, but one hopes no measurement haszero repeatability.Repeatability methodology is perhaps most useful in the context of quality assurancefor identifying both the source and magnitude of correctable error for groups ofscientific measurements before they are routinely used in future laboratory, greenhouse,and field trials. In May 1999 at IRRI, for example, a small repeatability studyof 20 participants who were asked to individually and independently record severalagronomic traits and injuries from the same 10 hills in the field ranked tiller numberas the most repeatable measure ( R = 0.8528 [mean of trained and untrained groups]),followed by plant height (0.7483) and number of whiteheads (0.5856; Schoenly andDomingo, unpublished data). In contrast, counts of tungro-diseased plants recordedby four observers at 10 sampling stations in one farmer’s field in Thailand (Disthaporn1987) gave an R value of 0.1407 and an upper 95% confidence limit of 0.5501. Withoutadditional repeatablility results at more sites, it is premature to ask whethermultiperson field counts of tungro-diseased plants are repeatable. relative to whiteheadcounts, for example. These studies, however, underscore the need for plant protectionworkers to conduct frequent repeatability trials of their scientific measurementsbefore they are put to routine use and to revise or even discard, if necessary,those measurements that are not repeatable.Other contexts where repeatability methodology is potentially useful include preandposttesting exercises in training workshops. This venue gives the principalinvestigator(s) the opportunity to observe the recorders firsthand and to detect andcorrect departures in protocol (Kahn and Sempos 1989). Publishing repeatability resultsalerts colleagues in plant protection disciplines to expected error values andconfidence limits for scientific measurements that are gathered under specific laboratory,greenhouse, and field conditions.82 Schoenly

ReferencesBecker WA. 1984. A manual of quantitative genetics. Pullman, Washington: Academic Enterprises.Disthaporn S. 1987. Studies on sampling methods for rice diseases in Thailand. Ph.D. dissertation.Justus-Liebig University.Kahn HA. Sempos CT. 1989. Statistical methods in epidemiology. New York: Oxford UniversityPress.Krebs CJ. 1989. Ecological methodology. New York: Harper-Collins.Lessells CM, Boag PT. 1987. Unrepeatabie repeatabilities: a common mistake. Auk 104:116–121.Ling KC. 1979. Variation in varietal reaction to rice tungro disease: possible causes. IRRI Res.Pap. Ser. 32:32–38.Sokal RS. Rohlf FJ. 1995, Biometry. 3rd edition. New York: WH Freeman and Company.Zar JH. 1984. Biostatistical analysis. 2nd edition. Englewood Cliffs. New Jersey: Prentice-Hall.NotesAuthor’s address: K.G. Schoenly, International Rice Research Institute. MCPO Box 3127.Makati City 1271, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL. editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management. 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Are tungro disease counts repeatable? 83

Surveillance scheme for tungro forecastingin MalaysiaA.B. Othman, M.J. Azizah, A.T. JatilRice tungro disease in Malaysia was suspected to occur first in 1933.Serious outbreaks occurred in 1982 and 1983, when more than 20,365and 12,439 ha, respectively, were affected. A nationwide campaign waslaunched to control the disease and to strengthen the rice pest surveillanceand forecasting system. The techniques used were mapping, fieldsurveillance, mobile nurseries, tests for viruliferous insects, and light traps.As a result of this campaign and the intensification of surveillance activities,the area of tungro infestation was reduced greatly. Green leafhopperresistantor moderately resistant varieties were widely recommended andadopted. Surveillance for tungro was further refined by incorporating newcomponents such as recording the severity of infestation, using a serologicaltest (ELISA) to detect infection, and using ultraviolet light traps formonitoring GLH abundance and species composition. Economic thresholdlevels were established for decision making. Other information on naturalcontrol, natural enemies, farmers’ attitudes and experiences, and availabilityof pest control equipment was also compiled.In Malaysia. tungro, or penyakit merah virus (PMV) as it is locally known, was suspectedto occur first in Kerian District in Perak State in 1933. At that time, the conditionwas thought to be caused by a physiological disorder, a soil problem, a nematode,or a combination of several factors. Later, however, it was reported to be causedby a virus (Ou 1965). For a long time, the occurrence of this disease was confined toKerian District, but in 1980 it was reported in Penang, Kedah, and Perlis. The mostserious outbreak occurred in 1982 in Kedah and Perlis when more than 20,300 ha ofrice fields were afflicted by tungro and yield loss was estimated to be 34,000 t, amountingto US$10 million. The 1982 outbreak has been attributed to several factors:• An increase in staggered planting, which resulted in the presence of a continuoussource of tungro.• A large population of active vectors, which created a greater potential for virustransmission.• The widespread cultivation of variety MR37, which was susceptible to both thevector and virus.• The availability of plant stages susceptible to the vector and virus.• Inadequate staff for pest surveillance and forecasting programs, resulting in ineffectivecoverage.Because tungro is disastrous, the Department of Agriculture (DOA) launched anationwide campaign that involved all agencies dealing with the control of the diseaseand strengthened the tungro surveillance and forecasting programs in Perlis,Kedah, Pulau Pinang, and Perak (DOA 1983). The surveillance techniques used duringthis time were mapping, field surveillance, mobile nurseries, testing for viruliferousinsects. and light traps.

As a result of this campaign and intensification of the surveillance activities, thearea affected by tungro in 1983 decreased to 12,439 ha. Subsequently, in 1984 and1985, only 3,843 and 978 ha were affected, respectively. From 1986 to 1989, the areainfected with tungro was between 330 and 740 ha. In 1990, however, 1,857 ha wereinfected with the disease (Chen and Othman 1991). This increase in tungro incidencewarranted quick action because rice planting was highly staggered due to problemswith water management and a high population of the major vector, Nephotettixvirescens. An intensive survey was carried out to determine the extent of the diseaseand infected plants and stubbles were destroyed by roguing or with herbicides. Insecticideswere used in areas with high vector numbers. GLH-resistant or moderatelyresistant varieties such as MR77, MR84, MR103, and MR 106 were widely recommendedand adopted. Subsequently, the affected area diminished greatly to only 401ha in 1991, 68 ha in 1992, 24 ha in 1993, 56 ha in 1994, and 22 ha in 1995. The tungrosurveillance program was then further refined by incorporating new components tointensify and strengthen surveillance activities and to continuously maintain tungroat very low or zero incidence. In 1996, the diseased area increased slightly to 3 17 ha,but the problem was contained by planting virus-resistant MR 159 in the infected areafor one season. As a result, no tungro was recorded in 1997 and 1998 (Fig. 1).Pest surveillance and forecasting systemThe pest surveillance and forecasting system in Malaysia was set up in 1979 by theDOA and implemented in all rice areas in the country. The merits of the surveillancesystem were fully realized in the light of the devastating outbreak of brown planthopperin Tanjung Karang in 1977 and mixed brown and whitebacked planthopper infestationsin Muda in 1978 and 1979. A project proposal with a budget of US$8.1 millionover a 5-yr period was approved by the Malaysian government and the DOA wasFig. 1. Annual incidence of rice tungro disease in Malaysia, 1983-96.86 Othman et al

entrusted with implementing the project. It was agreed that the Malaysian AgriculturalResearch and Development Institute (MARDl) would provide the necessaryresearch support. During the same period, the concept of integrated pest management(IPM) was adopted for rice in Malaysia and the pest surveillance and forecastingsystem became a crucial component. The system gathers information from the 450,000ha of rice fields in Peninsular Malaysia on the population of insect pests and naturalenemies, incidence and severity of rice diseases, infestation of major weeds, and damageby rats and golden apple snail (DOA 1989, Chang 1991). Other factors such ascrop age, water level, varieties, weather conditions, and farm practices are monitored.Light traps, net traps, mobile nurseries, and egg parasitism studies were set up toprovide additional information.Location of surveillance systemThe establishment of a surveillance and forecasting system involves setting up a practicalregional system in all the major rice-growing areas. Each regional system operatesfrom a pest surveillance center. The system is under an agricultural officer who isassisted by assistant agricultural officers, several agricultural technicians, and surveillancescouts. The number of agricultural technicians and scouts varies from onesurveillance center to another, depending on the size of the area. Ideally, each agriculturaltechnician supervises an area of 4,000–5,000 ha and is assisted by five trainedfield scouts. All the regional systems are coordinated by the pest surveillance andforecasting section based at the Crop Protection and Quarantine Services Divisionheadquarters in Kuala Lumpur. There are now 12 pest surveillance centers located inthe following areas: Tambung Tulang, Perlis; Telok Chengai and Sungai Petani, Kedah:Bumbong Lima, Pulau Pinang; Parit Buntar, Chenderong Balai, Seberang Perak, andTiti Gantung, Perak; Sungai Burong, Selangor; Gerai, and K. Terengganu, Terengganu:and Lundang, Kelantan.Surveillance techniques for tungroThe techniques adopted for tungro surveillance and forecasting are field surveys/fieldscouting, mobile nurseries, tests for viruliferous vectors, and light traps.Field surveillance/field scoutingIn field surveillance, the block surveillance system is used whereby several fieldswithin a 120–240-ha block are examined. All rice surveillance areas are divided intoblocks and coded accordingly by state, district, and locality. Ten fields per block arerandomly sampled and surveyed every 7–10 d. From each field, 10 hills are chosenrandomly and examined. Field scouts detect vector numbers and tungro disease incidence.Visual estimates of GLH are recorded. Visual counting permits rapid examinationwithout subsequent laboratory work, although accuracy probably decreases aspest density increases. At times, sweep nets are also used to estimate GLH numbers.Surveillance scheme 87

Initially, tungro incidence was assessed by symptoms and by the starch iodinetest. This test detects the increased starch content in rice leaves infected with tungro.The cut edges of infected leaves turn dark blue when tested. This test was abandoned,however, because other virus diseases such as dwarf disease and transitory yellowingas well as other stresses may also give similar results. Subsequently, rice leaves wereindexed for infection with tungro viruses by the latex flocculation test, which wascarried out at the surveillance centers.Presently, tungro field surveillance includes an assessment of the disease basedon severity of infestation. The index formulated for severity of infestation for tungrois 5%, 10%, 25%, and 50% diseased hills. A field manual has been developed fortungro surveillance and data management. Leaf samples are also collected for indexingfor tungro viruses by enzyme-linked immunosorbent assay (ELISA) at the surveillancecenters. Whenever required, the examination of rice stubbles for tungrosymptoms is also carried out and samples are taken for testing by ELISA. In high-riskareas for tungro, relatively more samples are collected during the field surveillance.For example, in Perak in 1998, more than 5,000 samples were collected for the ELISAtest. To complement field scouting, key farmers/farmer-leaders and surveillance brigadesthat consist mainly of farmers’ children have been trained to monitor tungroincidence and report to the surveillance staff. Farmers are encouraged to monitor thedisease in their own fields. Infected plants are removed and destroyed. All surveillancedata collected are analyzed on the same day.Mobile nurseriesMobile nurseries are deployed to detect the presence of any viruliferous GLH 1 mobefore planting. The approach involves planting TN1 (Taichung Native 1) rice seedlingsin trays, which are exposed to GLH in the field. The seedlings are exposed for 3days and nights. Ultraviolet lights or inflorescent lamps are used to attract GLH to therice seedlings in the tray. The tray is then brought back, sprayed with insecticide tokill the GLH, and placed in the glasshouse for 10-15 d. Initially, seedlings were assessedvisually for tungro and by using the starch iodine test. Seedlings are now testedserologically for each of the two tungro viruses (rice tungro bacilliform virus and ricetungro spherical virus) by ELISA. The surveillance centers that are equipped withELISA facilities are Telok Chengai, Parit Buntar, and Lundang, in addition to thePlant Disease Section in Kuala Lumpur. The number of mobile nurseries used dependson the size of the area monitored (Table l). If the ELISA test gives positiveresults and there is a high GLH population, the area and field extension officers areimmediately informed to allow further action to be taken. Intensive surveys and monitoringare carried out by the surveillance staff during the following season for up to60 d after planting.Transmission test for viruliferous vectorsGLH are collected from the field and used in transmission tests to determine theirinfectivity. After the introduction of ELISA, however, this technique is no longerused.88 Othman et al

Table 1. Number of mobile nursery trays (areas y -1 ).1996 1997 1998State Total trays Trays w/ Total trays Trays w/ Total trays Trays w/(no.) tungro a (no.) tungro a (no.) tungro a60Perlis1451188a Number of trays which contained tungro-infected seedlings.KedahPulau PinangPerakKelantan12220122061132340728018531018322454541245712143012915200Fig. 2. Monthly light trap catches of Nephotettix virescens (Nv) and N. nigropictus (Nn) in 1996-98in Kerian District, Malaysia.Light trapsLight traps are set up in the surveillance areas to detect the buildup of the GLH populationand its species composition. A high population of active vectors increases thepotential for virus transmission. In some areas, ultraviolet lights are used becausethey attract more GLH. Light trap catches are recorded daily. Intensive surveillanceactivities are carried out in areas previously infected with tungro when large numbersof GLH are caught in the light traps. For example, light trap catches in Kerian werelow in 1997 and 1998. This relates to the low or zero incidence of tungro infectionfound during the same period (Fig. 2). Constant monitoring of light trap catches hasenabled the early detection of tungro in rice areas and established a pattern of GLHactivities. The number of light traps in the major rice areas are Perlis, 5; Kedah, 20;Pulau Pinang, 4; Perak, 40; Selangor, 7; Terengganu, 6; and Kelantan, 9.Surveillance scheme 89

Processing surveillance dataData collected are examined and compiled on the same day. At the surveillance centers,data are studied by surveillance officers. Relevant information on the field situationis incorporated in the daily surveillance report. Data from light and net traps arealso reported as daily and weekly catches.Tentative economic thresholdsThe economic threshold is defined as the density at which control measures should beimplemented to prevent an increasing pest population from reaching the economicinjury level. The word “tentative” is used to describe the dynamic nature of the thresholdvalues, which are determined by variable factors such as yield, treatment cost, timetaken to initiate and carry out treatment, and perception of pest control by farmers. Tomake sound plant protection decisions, it is necessary to have tentative economicthresholds, however crude these may be initially. An arbitrary economic threshold fortungro is 1 GLH hill -1 (for areas with tungro) and for mobile nurseries 1% infection inELISA tests combined with a high GLH population.Decision makingHaving processed the surveillance data, the agricultural officers, assistant agriculturalofficers, and agricultural technicians involved discuss and assess the pest anddisease situation. In normal situations where pests and diseases are below thresholdlevels, farmers and extension officers are informed. This information is also displayedon notice boards at IPM clinics and meeting places for farmers and extension staff.Color-coded indicators are displayed at strategic places where farmers can easily seethem. The color codes used are green for safe, yellow for potentially reaching theeconomic threshold level (ETL), and red for above the ETL with action needed byfarmers.When pest or disease incidence approaches or exceeds the ETL, the informationis immediately relayed to the area extension officers and to the Pest Surveillance andForecasting Section in Kuala Lumpur. The extension officers and surveillance technicianswill then advise farmers and their children (surveillance brigades) to check theirown fields. An intensive survey is also carried out covering 25–50% of the blockarea.Other factors included in the data collected by field scouts, together with thetentative economic threshold, are natural control, farmers’attitudes, past experiences,and availability of pesticides and spraying equipment. At times, only spot spraying orspraying in certain fields is required. Should the pest or disease situation reach outbreaklevels in extensive areas, the pest control committee is activated. The committeecomprises members from the State Department of Agriculture, extension services,crop protection service, technical branches, and MARDI. In addition to reviewing thepest situation in relation to field conditions and making control decisions, the committeealso determines the surveillance for the period and ensures that decisions made90 Othman et al

are implemented immediately. The committee is also entrusted with planning pestcontrol when necessary. Any problems that need further research or technical problemsthat cannot be solved at the ground level are channeled to the IPM ImplementationCommittee in the state and to the National Rice IPM Technical Committee, ifrequired.Discussion and conclusionsThe success of a surveillance system can be measured by its ability to provide earlydetection of pests and to prevent a serious outbreak. In this respect, the surveillancesystem in the major rice-growing regions of Malaysia has been highly successful(DOA 1996, Chen and Jatil 1997). It has successfully contributed to the containmentof outbreaks of several pests and diseases through early warning and by aiding extensionpersonnel and relevant agencies to make decisions on control measures. Thesystem has helped to prevent GLH populations and tungro from reaching 1982-83levels. At present, the surveillance system for tungro very much depends on fieldscouting and mobile nurseries, which are time-consuming but very effective. Attemptsto improve field scouting are currently being carried out using ELISA. Antiserum forELISA is being produced for detecting rice tungro spherical virus (RTSV) and ricetungro bacilliform virus (RTBV) with the assistance of staff from IRRI and MARDI.This method would enable better supervision of the field staff because more fieldsamples can be verified. Preliminary observations showed that this method is notonly feasible but also more sensitive than visual assessment. The field surveillancedata in 1997 and 1998 showed no areas with tungro incidence, but samples analyzedby ELISA gave positive results. Symptoms were not expressed because plants wereinfected only with RTSV.Sometimes when symptoms are detected, it is often too late or beyond the actionlevel. What the tungro surveillance system needs is a supply of antiserum for RTSVand RTBV. This will allow the virus to be detected before symptoms are expressed.This approach will identify the areas infected with RTSV and RTBV, either with singleor double infections, to map the distribution and spread of tungro. Thus, effectivecontrol and eradication measures can be carried out. With this it is hoped that farmerswill be able to attain the potential yield of the crop.Light traps have been used to supplement field data. Light trap catches are knownto reflect the field situation of GLH incidence. Besides being used as a monitoringtool. light traps are sometimes used as a control measure. Natural enemies. particularlypredators of GLH, are recognized to be important mortality factors. Experiencein rice areas has shown that, where predators are common, no insecticides were used.The practice of herbicide application on bunds appears to be highly disruptive tonatural enemies of leaf feeders and leafhoppers. It is recommended that bunds beslashed mechanically to maintain refuges for natural enemies.Rice diseases in Malaysia are controlled whenever possible by incorporating hostplantresistance in MARDI’s rice varieties. Eighty percent of the rice planted is varietyMR84, which is moderately resistant to tungro and GLH-resistant. The use ofclean healthy seeds is a basic requirement in disease control. Field sanitation duringSurveillance scheme 91

and at the end of the crop cycle is a valuable cultural practice to reduce inoculumsources for the next crop. For tungro control, farmers are urged to destroy ratoon,stubbles, and volunteer seedlings to reduce the survival of hosts for GLH and tungro.The surveillance system allows tungro management programs to be evaluated effectively.Tungro is currently well under control and every effort is being made to maintainthis situation.ReferencesChang PM. 1991. An integrated crop management approach for major pests in broadcast direct-seededrice in Peninsular Malaysia. Paper presented at the Workshop on Surveillanceand Forecasting for Rice Pests, 23–25 April, Serdang. Malaysia.Chen MY, Jatil AT. 1997. Incidence and control of rice tungro disease in Malaysia. In: ChancellorTCB, Thresh JM, editors. Epidemiology and management of rice tungro disease.Chatham (UK): Natural Resources Institute. p 103–108.Chen MY, Othman AB. 1991. Tungro in Malaysia. MAPPS Newsl. 15(1).DOA (Department of Agriculture). 1983. Kempen kawalan penyakit merah di SemenanjungMalaysia. Laporan Suku Tahun, Cawangan Pemeliharaan Tanaman, Kuala Lumpur.DOA (Department of Agriculture). 1989. Workshop report on brown planthopper surveillanceand forecasting in MADA, Alor Setar, Kedah.DOA (Department of Agriculture). 1996. A field guide on surveillance of rice pests. Malaysia.Ou SH. 1965. Rice diseases of obscure nature in Tropical Asia with special reference to Mentekdisease in Indonesia. Int. Rice Com. Newsl. 15:4–10.NotesAuthors’ address: A.B. Othman, M.J. Azizah, and A.T. Jatil. Crop Protection Division, Departmentof Agriculture, Ministry of Agriculture, Jl. Gallagher. 50632 Kuala Lumpur, Malaysia.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1990 Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.92 Othman et al

Farmers’ rice tungro managementpractices in India and the PhilippinesH. Warburton. S. Villareal, and P. SubramanianRice tungro disease is a serious disease for farmers because it is difficultto forecast and control, and can cause high yield losses. In this study. wecompared the perceptions and practices of rice farmers in India and thePhilippines, two areas where tungro is reported as endemic. The aim wasto find out what farmers knew about tungro and how they coped with it. Wealso Investigated the factors that influenced their knowledge and managementpractices.Farmers and plant diseasesMuch has been written about farmers’ indigenous knowledge and there are manyexamples of farmers’ detailed knowledge of pests and crop protection methods (forexample. Boef et al 1993, Brammer 1980, Brokensha and Riley 1980, Fairhead 1993,Richards 1985). Virus diseases such as tungro, however, might pose problems forfarmers because the causal agent (virus) cannot be seen. Bentley (1992) points outthat farmers often know more about conspicuous and important pests (for example,weeds. grasshoppers, beetles) but less about inconspicuous pests. With a disease suchas tungro, which is difficult to observe but important in the damage that it causes,farmers’ understanding may differ considerably from that of scientists (Fig. 1). Ifresearchers are to develop better ways of managing tungro that are acceptable to farmers,they need to find out what farmers already, know about the disease and build fromthere.BackgroundThe two areas studied were Chengalpattu District, Tamil Nadu, India. and Midsayap,North Cotobato, Mindanao, Philippines. Both are considered tungro-endemic areasby the local agricultural research institutes.Focused group discussions and semistructured interviews were conducted initiallywith farmers to discuss their perceptions and practices relating to tungro diseaseand to gain insights into how farmers viewed the disease. This was followed by aseries of surveys of randomly selected farmers using a structured questionnaire. Theseconsisted of a baseline survey with questions on farming practices, pest and diseaseproblems, and knowledge and management of tungro disease, plus follow-up surveysto record actual farming practices for several seasons after the baseline survey (Table1). In the Philippines, the project had been established for a longer period so it waspossible to collect information from a wider range of villages. In addition, actualtungro disease incidence was monitored by researchers on 180 of the total sampleFarms, so it was possible to compare farmers' and researchers' observations of thedisease.

Fig. 1. Characteristics of four classes of farmer knowledge (from Bentley 1991, Fig. 2.).Table 1. Data collection strategies used.IndiaChengalpattu District, Tamil NaduGroup interviews in 6 villagesQuestionnaire surveys of 90 farmers in 2 villages:baseline plus follow-up surveys for 6 seasons,1996-98PhilippinesMidsayap, North Cotabato, MindanaoGroup interviews in 9 villagesQuestionnaire surveys of 226 farmers in 9villages: baseline plus follow-up surveys for4 seasons, 1996-97Table 2. Farming systems used in India and the Philippines.Item India PhilippinesCrops Rice, sugarcane, groundnut RiceAv. no. rice crops year-1 2.3 2.0Irrigation system Tank, tube well Gravity (National IrrigationAdministration)Farm size (ha) 2.78 (std. 3.22) 1.31 (std. 0.98)Main planting method Transplanted Direct-seeded (63.5%)ResultsIn both areas, rice was the major staple crop, and multiple crops of rice were grownper year (Table 2); however, India has a more mixed cropping system than the Philippines,with other crops such as sugarcane and groundnut rotated with rice. The irrigationsystem also differed, with the Filipino farmers relying on a large-scale gravitysystem, whereas Indian farmers used tanks or wells. India has three distinct seasons(sornavari, samba, and navarai); the Philippines has two main seasons, dry and wet,but farmers can plant rice year-round. The majority of farmers had small farm sizes,with 87% of Indian farmers and 99% of Filipino farmers having 5 ha or less.94 Warburton et al

Experience with tungroAlthough both areas were identified as tungro-endemic, Filipino farmers reported ahigher incidence of the disease; 85% of the farmers had experienced tungro at leastonce, and more than 40% had experienced it two or more times. Seventy-five percentof Indian farmers had experienced tungro at least once, but few could recall morethan one disease attack (Table 3). In fact, Indian farmers could only recall one occasion(1990-91) when villages had been badly affected by tungro. In the 1996-98 seasons,tungro incidence in Philippine villages was again much higher, with some incidencefound in every season (Table 3). In Indian villages. tungro only occurred in onevillage in one season (samba 1997) (Table 3). Although the data are based on researchers’measurements in the Philippines and farmers’ reports in India, incidence isundoubtedly much higher in the Philippines.Importance of tungro relative to other pests and diseasesNot surprisingly, Filipino farmers rate tungro as a far more important pest or diseasethan Indian farmers (Table 4). Indian farmers generally appear to be less “pest-conscious”than Filipino farmers, with many stating that they did not have major pest ordisease problems in many seasons (particularly in the dry sornavari season ).Table 3. Farmers’ experience of tungro disease.Percentage of farmers who reported experiencing tungro prior to 1996No. of tungro attacks India (n = 90) Philippines (n = 226)experienced01234525.666.77.800014.243.420.413.30.40.4>50 8.0India: Percentage of farmers reporting tungro incidence, 1996-98 (n = 90)96 Samba a 97 Navarai 97 Sornavari 97 Samba 98 Navarai 98 SornavariTungro 0 0 0 7.8 0 0No tungro 100 100 100 92.2 100 100Philippines: Percentage of farms where tungro incidence was recorded by researchers, 1996-97Tungro incidence0=< 5%=< 50%> 50%Farms (no.)96 WS18.759.120.41.822597 DS28.564.55.21.717297 WS42.750.65.41.323996-97 WDS31.058.19.61.3229a Samba = monsoon crop/July-December planting, navarai = low rainfall crop/December-January planting, sornavari= May-early June planting.Farmers’ rice tungro management 95

Table 4. Most important pests and diseases reported by farmers in baseline survey,1996.lndiaSornavari Samba NavaraiStem borerLeaffolderNeck blastBPH a hopperburnEar head bugStem borerLeaffolderTungroCutwormLeaffolderStem borerEarhead bugBlastPhilippinesDry season Points b Wet season PointsStem borerTungroBPH/hopperburnBlack bugRatRice bugaGLHCutwormWormGolden snailArmywormLeaffolderNeck rotOthers29125520418912968473729191212727Stem borerTungroBPH / hopperburnBlack bugRatGLHRice bugCutwormWormGolden snailLeaffolderArmywormNeck rotOthers27727518918313962553127191412831a BPH = brown planthopper, GLH = greenleafhopper.b Farmers’ rankings on a scale of 1st = 3 points,2nd = 2 points, and 3rd = 1 point.Knowledge of tungro symptomsAll farmers in the Philippines knew of tungro, but 21% of Indian farmers had no ideawhat it was (Table 5). Farmers had not heard of tungro or its local name. Other farmerswere able to give a reasonable description of the disease, with 37% and 50% offarmers describing both the yellowing leaves and stunted appearance in India and thePhilippines, respectively. Farmers described tungro as similar to cancer or AIDS becausethey knew that the plant, once infected, would not recover. In the Philippines,the word “tungro” was used generally to describe a devastating problem, but farmersdid not use it to cover any type of damaging rice disease. They could name otherconditions and diseases of rice that produced similar symptoms, but that they knewwere not tungro (e.g., zinc or nitrogen deficiency). Some researchers observed thatsome farmers do have difficulty, however, in distinguishing tungro from zinc deficiency.Farmers’ perceptions of causes and mode of tungro spreadFarmers were uncertain about the causes and mode of spread of tungro (Table 6). Halfthe Indian farmers and 14% of Filipino farmers said they had no idea about the causeof tungro. Insects were identified most often in India as the cause of tungro spread,96 Warburton et al

Table 5. Farmers (%) reporting tungro symptoms.Symptom lndia (n = 90) Philippines (n = 226)Yellowing leavesStunted appearanceYellowing leaves & stuntinglncompatlble symptomsDo not knowHave not heard of tungro11.117.836.7013.321.145.1050.40.44.00Table 6. Farmers (%) reporting causes of tungro. Some farmers specifiedmore than one cause or mode of spread.Factor India (n = 90) Philippines (n = 226)Green leafhopperBrown planthopperinsects (general)Other specific insectsWaterWind/airSoil/rootsWeatherSeedsBirdsSusceptible varietyOld rice varietiesContinuous croppingVirus, germNo idea5.6036.7002.200000003.350.036.315.05.34.917.37.14.95.33.50.420.30.40.415.914.2and more than a third of Philippine farmers knew that tungro was spread b) greenleafhoppers. Only 6% of Indian farmers knew this. Other modes of spread such asthrough water, air, and soil were also identified.Filipino farmers were more aware of the effects of varietal selection on tungroincidence. Through their own observations, they had noticed that some varieties weremore susceptible than others to tungro.Farmer management strategiesFrom a researcher’s point of view, management of tungro is based on controlling thesources of disease inoculum or controlling the disease vector, the green leafhopper.Methods found by researchers to be effective include using resistant rice varieties andusing synchronous planting, rotations, or fallow periods to remove rice plants at certaintimes of the year, thus removing sources of disease inoculum and preventingdisease spread. Other methods involve roguing diseased plants and applying fertilizer.Insecticides used to be widely recommended for controlling the green leafhopper,although some researchers doubt their effectiveness.Farmers' rice tungro management 97

In the farmers’ baseline surveys, farmers were asked what they would do to managetungro. In addition, actual farming practices that might affect or be affected bytungro, such as choice of variety and insecticide use, were also recorded over severalseasons.Farmers’ views on how to manage tungro diseaseA significant number of Indian farmers (43%) did not know how to manage tungro.More than half (52%) of the farmers, however, suggested leaving a fallow period(Table 7). (Some said this was recommended by extension people.) Only 2% suggestedusing insecticides.In the Philippines, insecticide use was more popular, particularly at the earlystages of crop growth. Farmers realized that spraying was not effective at the latestages once the crop was already infected. Many farmers also said that insecticidespraying was not very effective at any stage, but that they used it in the absence ofother effective measures.Filipino farmers frequently suggested using resistant varieties or changing varietiesto prevent recurrence of tungro. Cultural controls such as roguing infected plants,fallows, plowing under, and synchronous planting were not commonly used (Table7).Farmers’ actual farming practicesThe rice variety and the amount of pesticide spray used by farmers (and reasons forthis) were monitored over a period of 6 and 4 seasons in India and the Philippines,respectively. Information on synchronous planting and rotations was also collected.Rice varieties used. In India, farmers chose varieties they considered as suitablefor the particular season in terms of characteristics such as growth duration, tolerancefor cold, and tolerance for pests. For example, IR50 is very popular in the dry, warmsornavari season when there are relatively few pests, but it is not considered so suitablein the wetter or colder seasons.IR50 was one of the varieties affected by tungro in 1990-91 in the area, but itcontinues to remain popular with farmers. Besides IR50, other varieties known to besusceptible to tungro are IR36, ADT36, IR64 and, particularly, ADT42. During anoutbreak of tungro in one village (Vishar) in the 1997 samba season, 7 out of 10farmers affected were growing ADT42; the others were growing IR36 or ADT36. Inthe other unaffected village (Pudumanvilangai), few were growing ADT42. In thefollowing navarai season, fanners switched from ADT42 and many planted ADT37,which has some tungro resistance. Tungro did not recur.In the Philippines, it was difficult to determine exactly what variety farmers weregrowing because they are very active in selecting their own seed. Selection was themain variety grown although varieties differed from farmer to farmer (Table 8). Farmersoften gave local names to their varieties and it was difficult for researchers to find outthe original rice variety. Farmers were aware of which varieties are susceptible to98 Warburton et al

Table 7. Farmers’ reported tungro management strategies.lndiaControl measure Farmers (%) using control in = 901Spray insecticidesPractice field sanitation & sprayingUse resistant varietyUse fallow periodfallow & change varietyNo idea2.21.11.152.21.143.3PhilippinesControl measure Farmers (%) using control (n = 226)When tungro To control tungro To control tungro To preventoccurs in nearby at early stages of at late stages of recurrence infield crop crop next seasonDo nothingApply insecticidesUse resistantvarietiesChange varietyRogue Infected plantsReplowAdd fertilizerPrepare land& sanitizeCheck water control/drainagePractice synchronous/early plantingObserve fallow periodPlow underPractice crop rotationFollow other controlmeasuresAsk adviceDo not know/no answer9.366.83.52.20.90.42.70.98.00.400.40.40.42.29.34.453.500.48.023.912.805.800000.91.38.839.844.7000.91.34.003.100000.91.88.41.85.347.311.1001.314.22.71.82.20.40.41.30.99.3tungro, and a significant number were growing varieties with some resistance (suchas IR62 and IR74). A few farmers were growing varieties such as the very susceptibleIR64; some farmers liked to grow special varieties such as Masipag, despite its susceptibilityto tungro.Insecticide use. The level of insecticide use was far lower in India than in thePhilippines, averaging between one and two applications in India and four to five inthe Philippines (Table 9), although extension documents in India generally recommendmore insecticide use for pest control than do such documents in the Philippines.The high level of insecticide use in the Philippines, however, was due in large part toFarmers’ rice tungro management 99

Table 8. Rice varieties grown by farmers.India: Farmers (%) using each variety averaged over 3 yrVariety Sornavari Samba NavaraiIR50 75.5 5.2 5.6IR36 b 5.5 10.4 10.4ADT36 b 4.6 11.8 23.3ADT37 a 2.1 8.1 13.3ADT42 b 1.8 7.8 4.4IET1444 0.9 0 6.7White ponni 0.3 12.2 1.5ADT39 0 4.8 1.5ENT2 0 3.3 4.1IR64 b 0 2.6 1.5Other varieties 1.2 3.0 3.3No rice crop 10.9 34.4 29.6Philippines: Farmers (%) using each variety in 1996Variety Dry season Wet seasonSelectionIR62 aIR36Bordagol7 Toner bIR78IR60PSB Rc10KoreanIR66Masipag bCaliforniaPSB Rc18 aIR88555IR74 aSeriesB6Others27.111.59.97.34.74.24.24.24.24.22.11.61.61.01.01.00.50.59.423.88.96.08.52.46.94.84.03.22.43.61.61.22.01.61.22.82.012.1a Carries some resistance to tungro.b = susceptible to tungro.the arrival of black bug in the rice-growing area. This new and very visible pest is ofenormous concern to farmers, who are not used to dealing with it. A detailed analysisof why farmers sprayed insecticide (Table 10) indicated that black bug was their mainreason for spraying. Tungro accounted for less than 5% of the total sprays. There wasno correlation between the number of insecticide sprays and tungro incidence.100 Warburton et al

Table 9. Mean number (standard deviation) of insecticide applications applied over 6 seasons inIndia and over 4 seasons in the Philippines.lndia96 Samba 97 Navarai 97 Sornavari 97 Samba 98 Navarai 98 Sornavari1.0 (0.74) 1.0 (0.89) 0.9 (0.79) 1.6 (1.03) 1.4 (0.90) 1.6 (0.94)Philippines96 WS 97 DS 97 WS 96-97 WDS4.6 (2.52) 5.0 (2.96) 4.2 (2.36) 4.5 (2.91)Table 10. Reasons of farmers for spraying insecticides (percentage of totalsprays), Philippines.Pest 96 WS 97 DS 97 WS 96-97 WDSBlack bugAny pest/for protectionStemborerGLHBPH/hopperburnTungroRice bugWormsLeaf folderOtherTotal spraysTotal parcels53.325.312.86.05.04.24.13.71.00.51,39328958.235.89.21.32.31.00.91.00.2098818626.856.74.07.02.83.81.75.300.51,10024470.919.710.73.72.41.12.01.60.30.41,161249Note: Multiple answers.Planting dates and other cultural controlsMore synchrony in planting dates occurred among Indian farmers than among Filipinofarmers. In the Philippines, it was difficult to determine the cutoff point betweenone cropping season and the next. For example, planting dates for the 1996 wet seasonvaried from 5 April to 24 August; for the 1996 late wet season, planting dateswere from 10 August 1996 to 30 January 1997; and for the 1998 dry season, from 18December 1997 to 30 May 1998. The irrigation water schedule was a major factor infarmers’ choice of planting dates.Indian farmers also used more rotation systems; for example, rice-rice-groundnutor sugarcane for 2 yr, then rice. Many farmed a mix of irrigated, partially irrigated,and rainfed land. There were more rotation and fallow systems inPudumavilangai (the village with one reported tungro occurrence) than in Vishar (thevillage where a tungro outbreak occurred in 1997).Farmers’ rice tungro management 101

DiscussionFactors affecting farmers’ perceptions and management practices of tungro includedcharacteristics of the disease, farmers’ access to information. plus their own observationsand experiments. In addition, farmers’ social and economic status and their wholefarming and livelihood system affected the management strategies that they can implement.Tungro is difficult to manage because it is not easily predictable. In India, farmersactually had little experience with it; therefore, their knowledge was more limitedthan farmers in the Philippines. They had not had the chance to experiment withdifferent management strategies and observe the outcome. They were also much lessconcerned about the disease because it did not recur season after season, possiblybecause their farming systems were more mixed. Farmers’ information sources forthe outbreak in 1990–91 were mainly extension people, who advised some of them topractice fallowing.Information about differences in varietal susceptibility is something that farmersseem quick to pick up and act on. In the outbreak in one Indian village in 1997,farmers were able to observe the differences between susceptible variety ADT42 andresistant variety ADT37. In the following season and following year, farmers droppedADT42 and many planted ADT37 (T. Chancellor, personal communication 1998). Inthe Philippines, farmers are very active in selecting for seeds that they think will givegood yield. The speed with which farmers adopt varieties was illustrated by the outcomeof an on-farm experiment conducted by researchers at IRRI and PhilRice. Thistrial was designed to compare a new tungro-resistant line with other varieties. Thefarmer-collaborator observed the results, selected seed from the new line, multipliedit, rechristened it with a new local name, and then distributed it among other farmers.Farmers, however, do not always fully understand the implications of plant resistance.For example, some believe that tungro disease is in the seeds, or they wonderwhy a particular variety does well in one field but becomes infected in another field.Although this is a difficult issue, because host resistance can break down over time orin different locations, there is a need for more and clearer information on tungroresistantvarieties for farmers.Farmers do not generally understand the risks of having infected rice plants aroundas sources of inoculum. In less intensively cropped areas where other crops can begrown, this is not a big problem. But with asynchronous planting where there is alwaysrice in the vicinity and limited options (because of water control) for alternativecrops, this poses a problem for tungro management. Only if farmers realize the potentialrisks of having infected rice in the field will they have any rationale to aim forsynchronous planting or rotation/fallow periods.In the Philippines, using synchronous planting, fallows, and rotations is likely tobe a greater challenge than in Indian villages. These options require control overwater, which, in turn, requires negotiations with irrigation authorities and neighboringfarmers. Alternative crops (for a rotation system) may not be possible if fields arepoorly drained, and Philippine farmers have fewer alternative sources of income (other102 Warburton et al

than rice), so they may be reluctant to use fallows. Indian villages have greater localcontrol over water supply, and the farming system is more mixed.The use of resistant varieties is a more straightforward method for farmers toadopt than cultural controls because it does not require group action or agreement.Farmers given training and information on tungro management including the use ofcultural controls become aware of the problems of asynchronous planting and tungroincidence, and can try to minimize tungro incidence.Insecticide use depends on other pests observed by the farmer as well as tungro.Although insecticides are commonly used, particularly in the Philippines, there doesseem to be a realization that insecticides are not the complete answer to tungro andcannot cure the disease. Confusion over the tungro vector, the green leafhopper, impliesthat insecticide application is unlikely to be closely targeted at the vector.ConclusionsIt is apparent that farmers have uneven knowledge of tungro, especially about thevector and sources of inoculum. They can adopt some control measures, however,such as the use of resistant varieties, without having a detailed understanding of thecauses of the disease. For management strategies that are difficult to adopt, such aschanging planting dates or using fallows, farmers need to understand why these methodscould be advantageous. Otherwise they are unlikely to put time and effort intotrying them out.For researchers, it is useful to be aware of what farmers already know. Researchersneed to provide relevant information, for example, on resistant varieties, to enablefarmers to make informed decisions. Information on the types of virus involved, howeverimportant to researchers, is much less useful to farmers because it does not havea direct impact on their management strategies. It may also be usseful to reconsiderwhat is meant by “tungro-endemic” area. In the case of India. labeling villages astungro-endemic (a term applied to the whole district) does not seem appropriate inguiding control recommendations, given the low frequency of tungro outbreaks.ReferencesBentley JW. 1992. The epistomology of plant protection: Honduran campesino knowledge ofpests and natural enemies and the implication for control strategies. Proceedings of theCTA/NRI Seminar on Crop Protection for Resource-Poor Farmers. 4–8 November 1991.University of Sussex. Chatham, (UK): Natural Resources InstituteBoef W de. Amanaor K, Wellard K, Bebbington A, editors. 1993. Cultivating Knowledge: geneticdiversity, farmer experimentation and crop research. London: Intermediate TechnologyPublications Ltd.Brammer H. 1980. Some innovations don't wait for experts. Ceres 132. March-April. p 21–28.Brokensha D. Riley BW. 1980. Mbeere knowledge of their vegetation. In: Brokensha DW,Warren DM. Werner O, editors. Indigenous knowledge systems and development. Maryland:University Press of America.Farmers’ rice tungro management 103

Fairhead J. 1992. Indigenous technical knowledge and natural resource management in sub-Saharan Africa: a critical overview. Paper presented at the Social Science Research CouncilProject on African Agriculture Conference, January 1992, Dakar, Senegal.Richards P. 1985. Indigenous agricultural revolution: ecology and food production in WestAfrica. London: Hutchinson, and Boulder, Colorado: Westview Press.NotesAuthors’ addresses: H. Warburton, Natural Resources Institute, University of Greenwich, CentralAvenue, Chatham Maritime, Chatham, Kent ME4 4TB, UK; S. Villareal, InternationalRice Research Institute, MCPO Box 3127, Makati City 1271, Philippines; P. Subramanian,Tamil Nadu Agricultural University, Coimbatore 641003, Tamil Nadu, India.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.104 Warburton et al

Community-based rice pest managementX.H. Truong, E.H. Batay-an, S.C. Mancao, G.N.A. Abrigo, A.B. Estoy, L.B. Flor, Jr., H.D. Justo, Jr.,E.R. Tiongco, R.N. Casco, and S.R. ObienAn exploratory interdisciplinary approach with farmers’ participation gavean expected and much-needed boost to community efforts to manage pestproblems in three villages in Midsayap, North Cotabato. This simple interventioncould slowly halt emerging pests to the area. A strategic actionplan prepared after informal group dialogue was designed for farmers tohelp them understand the rationale of the project, learn from their farmingexperiences from the crisis due to damage caused by rice black bug (RBB)and rice tungro disease (RTD) in 1996, and alleviate fears of threats totheir income, while sustaining their rice production. The plan encouragedfarmers to participate in establishing their crop simultaneously with theirneighbors, choose suitable varieties and crop establishment methods, andmonitor pests prior to taking appropriate management action. The communityshared ideas and experiences on crop management through a consultativeand planning workshop, demonstration, experiments, and a farmers’field day. Based on the participatory rural appraisal method, theirindigenous knowledge of farming, pest and control practices, financial constraints,and the irrigation schedule was used to establish an action plan.A total of 140 farmers in Bual Norte, Bual Sur, and Bobonao villages inMidsayap, North Cotabato, participated in the project for two to four croppingseasons in 1996-98. Most farmers (82–97%) realized that the intensiveuse of Insecticides could not control the major pest problems of RBB,RTD, and white stem borer (WSB) in the staggered planting system. Theychanged gradually from staggered to synchronous planting with a 45–56-dand 30–45-d fallow period in each village. They shifted from transplantingto the broadcast method of planting (76–82%) and selected early maturingvarieties (65–79%) and recommended varieties (20–35%). With this practice,farmers in these communities obtained an average rice yield of 3.2-4.6 t ha -1 and reduced RBB density (1.3–2.0 hill -1 ), RTD-infected farms(2.2–11.5%), and whitehead incidence caused by WSB (3–22%). They reducedthe frequency of insecticide application from 5–6 times to 3–3.4times per season. In general, farmers’ perceptions and experiences Influencetheir cultural management and pest control practices, which are alsoclosely related to their sociodemographic characteristics. Thus, most farmerscontinued searching for varieties resistant to pests through seed exchangeswith their friends. A few farmers still used insecticides intensively,especially those with an average income level. Some were aware of thepresence of spiders and parasitized egg masses of WSB through informationthey received, but most farmers have not translated their ideas intofarming practices. The results clearly suggest that the project’s strategicaction plan in the future should focus on the needs of specific target farmergroups.Rice tungro disease (RTD) has long been considered as one of the most importantfactors limiting rice production in the irrigated lowland ecosystem, notably insouthern Philippines. Outbreaks sporadically occurred in locations commonly associatedwith staggered planting practices. Besides RTD, farmers in this ecosystem havesuffered huge economic losses caused by complex pest problems such as rice black

ug (RBB), Scotinophara coarctata (Fabricius), and white stem borer, Scirpophagainnotata (Walker). To address farmers’ pest concerns, (1) information on rice tungrovirus disease (Tiongco 1996) and integrated management of RBB (Justo 1995, and(2) promising technologies have been promoted through training for extension workersand farmers. Most activities were implemented by single or multidisciplinary teamswith farmer participation as a basic step leading to pest management. However, sucha research approach focused more on pests, in contrast to farmers’ pest control practices,which are based on their perceptions, experiences, and socioeconomic resources.As agricultural research becomes increasingly specialized, there is a great need tointegrate information across disciplines or among agencies to develop technologiesfor the total farming system, which is seen as a major challenge.This report details how an interdisciplinary approach was used to integrate researchoptions and indigenous knowledge (IK) of farmers from three villages aroundthe PhilRice Branch Station at Midsayap, North Cotabato, Mindanao. It has the followingobjectives:1. To assess farmers’ perceptions and knowledge of pest control practices in ricecultivation as well as constraints in their indigenous rice cultivation,2. To share with farmers research options that answer their needs to strengthen theirdecision making in managing the rice crop and pests and3. To facilitate farmers’ participation in sharing ideas and experiences in the communityand on their farms.MethodologyEstablishment of baseline survey informationStudy sites were selected based on the pest profile of Mindanao reported earlier bySanchez and Obien (1995). The social science group began a baseline survey to gatherinformation on farmers’ perceptions and knowledge of pest control practices. Brainstorming,planning, and orientation among team members across disciplines weredone at the study site to determine information needs and how they could be bestobtained. These activities were also used to generate a list of potential participants inthe community survey from village officials, farmer-leaders, and key informants. Onehundred farmers or about 20% of the farmer population in Bantod and Bual Sur werechosen as respondents.Interagency planning workshopAfter gathering information from farmers, a planning workshop was held at the PhilRiceBranch Station at Midsayap. Representatives of different agencies under the Departmentof Agriculture (DA), farmers, cooperatives, community council members andleaders, chemical sales representatives, and municipal agricultural officers at the studysites were invited to discuss and contribute to the action plan and future expansion ofthe project. They were also asked to examine the performance of newly recommendedvarieties in the production experiment.106 Truong et al

Establishment of farmers’ indigenous knowledge and participationThe initial phase of the participatory rural appraisal (PRA) consisted of visits withvillage officials. farmer-leaders, and key informants. To enhance community participation,the group decided to use the informal dialogue and discussion forum to gatherIK on rice cultural practices. Based on the workshop information, questionnaires wereredesigned to come up with a conceptual framework. The framework represents acontinuing cycle focused on incorporating IK into decision making for sustainablerice production and community pest management (Fig. 1). Seven farmer groups participatedin the 5-d survey. Participants were grouped and the discussion covered riceproduction, the pest scenario, and the project plan.The last phase of the PRA collected IK through informal focused interviews andcreated a venue for community awareness and participation in implementing the project(Fig. 2). The sociodemographic profile of respondents in relation to their rice culturalpractices was established and analyzed by factor analysis (Statistica).Community-based rice pest management 107

Fig. 2. Diagram of the project implementation strategy in three villages, Bual Norte, Bual Sur, andBobonao, Midsayap, North Cotabato (1996-98). LGU = local government unit, MAO = municipalagriculture office, DA = Department of Agriculture, NIA = National Irrigation Administration, PRA =participatory rural appraisal.Verification trials of farmers’ IKFarmers’ selected varieties and the direct seeding method, two IK components, werevalidated in a series of demonstration experiments at the research station. Cooperatorand noncooperator farmers were invited to assess varietal performance and evaluatethe treatments. Cooperators were offered 10 kg of seeds of their chosen varieties withinformation on varietal characteristics and performance, and responses to insects anddiseases in different rice ecosystems (PhilRice Technoguide Calendar 1997-98).Monitoring of crop performance, diseases, insects, and natural enemiesRegular monitoring of the green leafhopper, Nephotettix virescens; whitebackedplanthopper (WBPH), Sogatella furcifera; zigzag leafhopper (ZLH), Recilia dorsalis;white stem borer (WSB,) RBB; and mind bug (MRB), Cyrtorhinus lividipennis,has been carried out by using a light trap (adapted IRRI-EPPD design with a 160-Wmercury lamp) set up at the PhilRice Experiment Station at Midsayap since 1997.108 Truong et al

Meanwhile, insects, diseases, and natural enemies were monitored by insect netsweeping from three quadrats (2 × 5 m each) taken from farmers’ fields and anotherthree from a 1000-m 2 farmer’s field at 30 and 45 d after planting (DAP). Yields wereevaluated based on the cut crop from those quadrats.Cooperators were also provided with fertilizer (NPK) enough for a 1000-m2 experiment.The effect of zinc deficiency was monitored during the rainy season todecide on an amendment. A 1-d field visit (farmers’ field day) took place at cropmaturity. where farmers were asked to share their experiences. Substantial componentsof rice cultural practices related to crop yields in each of the three farmercommunitieswere determined by the principal component analysis (SYSTAT).Continuing campaign for community synchronous plantingAn information campaign was launched to develop awareness about communitycooperation in regular planting according to the schedule of water release in the villageand fallow period to reduce pest pressure during land preparation; farmers’ fielddays; field visits; and pest monitoring. Increased knowledge was one of the ultimatetargets of the project activities.Results and discussionBaseline survey informationCommunity sociodemographic profile. Table 1 summarizes the profile of 100 farmerrespondents(74 men and 26 women) from two villages, Bual Norte and Bual Sur,Midsayap. Most farmers (82%) obtained their income from rice farming. Respondentsgot loans for production inputs from traders (21%), private agents (18%), cooperatives(15%), and landowners (1%). Only 38% of the farmers were landowners.One young farmer with a college education and three old farmers with a high schooleducation were selected by the community as farmer-irrigators responsible for communicatingwith the National Irrigation Administration (NIA) to set the water releaseschedule in their areas. Because of the insufficient irrigation water in Midsayap, particularlyduring the dry season, NIA recently launched a new irrigation program forthe different zones or divisions composed of several villages. The specific week ormonth of water release, however, depended on the preference of farmer-irrigators inthe village. This initiative of NIA encouraged farmers to practice synchronous planting.It benefited both farmers and NIA because NIA was able to collect irrigation feesfrom farmers who produced more than 2 t ha -1 of rice per season. The limitation of theprogram was that frequently two nearby villages did not have water at the same time;thus, synchrony cannot be practiced in large contiguous areas.Farmers’ perceptions on insect pests and diseases. Rice tungro was perceived asthe most important disease of the rice crop; bacterial leaf blight (BLB) or blast rankedsecond (Table 2). Crop loss due to RTD ranged from 0% to 91% based on farmers’past experiences. Farmers associated the spread of the disease with farming or weatherfactors rather than with the disease causal factor. The use of insecticides and GLH-Community-based rice pest management 109

Table 1. General profile of farmer-respondents (100) in Bual Norte andBual Sur, Midsayap, North Cotabato, 1996.General featuresArea covered (ha)Average farm size (ha)Source of irrigation (%)Sociodemographic characteristicsAverage age (yr)Yr farmingSex (%)MaleFemaleCivil status (%)MarriedSingleHousehold members (no.)Formal education (%)Membership in organizations (%)Farmers’ associationFarmers’ cooperativeIrrigationNoneNo schoolingElementaryHigh schoolCollegeSocioeconomic characteristics (%)Tenure statusownerCertificate of land transferTenantLesseeSource of financial input OwnTraderAgentCooperativeLandownerFarming equipment/animal Carabao, owned (%)Tractor, ownedThresher, rentedSource of incomeRice farmingRice-cum-mungbeanAnimal raisingBusinessGovernment employeeValue129.71.310044.521.874.026.096.04.06.04.047.042.07.032.04.03.058.037.533.722.16.745.120.618.314.71.050.06.0100.082.50.86.73.36.7resistant varieties were the preventive measures applied. Thirty-eight percent of thefarmers claimed that insecticide use was an effective control measure, while 31%said otherwise. A few farmers (5%) confused RTD symptoms with zinc deficiency. Amajority of the farmers (71%) used inappropriate control measures for BLB.Meanwhile, 44% and 20% of the farmers considered RBB and WSB as the firstand second major pest concerns, respectively (Table 3). They correctly identified signsof WSB and RBB damage. Crop loss caused by RBB ranged from 0% to 88%, and110 Truong et al

Table 2. Most important diseases for farmers.Disease Response (%)Rice tungroPerception of causes and spread of rice tungro (n = 45)Bad weather/rain/stagnant waterInsects“Tiny creature”Pests are contagiousOverfertilizationOld varietyThrough seedDo not knowPerception of crop loss (%) due to rice tungro (n = 55)07-20>20-40>40-60>60-91Respondents’ preventive practices for rice tungroUse insecticidesDram + add zincLeave itUse water + spray + apply fertilizerNo responseEffectiveYesNoYesNoNoNoNot applicableBacterial leaf blight (BLB)/blastPerception of crop loss (%) due to BLB/blastWith BLB/blast (n = 35)>0-20>20-40>40-60>60Without blast/BLB (n = 65) Not applicablePerception of causal factors of BLB/blast (n = 35)CausesPestsFertilizer +/-Do not knowBad weatherDisease is air-borneNo responseControl practices for BLB/blast (n = 35)20.018.216.416.61.81.81.824.412.712.729.129.19.138.230.95.41.83.69.110.942.937.114.35.725.725.722.911.42.911.4Drain + use insecticides 71.4Drain + apply fungicide(hinosan) + add fertilizer 14.3Add fertilizer 5.7No response 8.6Community-based rice pest management 111

Table 3. Pests ranked by farmer-respondents as first and second, andfarmers’ control practices.Pest or practice Crop loss (%)Rank 1Rice black bug (RBB) (n = 44)0-20>20-40>40-60>60-88White stem borer (WSB) (n = 20) 14-20>20-40>40-65Hoppers/hopperburn (n = 4) 42.9Rats (n = 3) 48.1Whorl maggot (n = 1) 29.2Cutworm (n = 1) 28.6Not concerned (n = 27)Not applicableResponse (%)81617331164311Rank 2White stem borer (WSB) (n = 54) 08-20>20-40>40-50Rice black bug (n = 10) 0-10>10-20>20-60Hoppers/hopperburn (n = 16) 10-20220-40>40-60Whorl maggot (n = 1) 20No infestation (n = 19)Not applicable5222254429251Control practices against RBB and WSBPractices of respondents who observedpests (94%)Used pesticides (99%)Apply insecticideSpray/adjust timingIrrigate + sprayBait + give poisonDid not use pesticides (1%) Irrigate + fertilizerRespondents who did notNot applicableobserve pests (6%)45.745.76.41.11.1that caused by WSB from 14% to 65%. Most farmers experienced WSB damage atthe vegetative (deadheart) and reproductive stages (whitehead). Some also receivedinformation on monitoring of parasitized egg masses, but they said that the activity istime-consuming. The 1996 dry season was the first time in 30 yr, however, that theyexperienced an RBB outbreak.112 Truong et al

Farmers’ knowledge of pest control measures. One-half of the respondents hadattended seminars on topics related to farming in general, with 40% of the topicsfocused on pest management (Table 4). The most common information, however,was on how to use pesticides. They received this information from DA technicians(35%) and by radio (11%). The DA (26%) also advised them to monitor insect pestsand natural enemies. Water management combined with proper timing of insecticideapplication was the most common pest control alternative against RBB cited by respondents.whether they attended the seminar or not (23% and 21%, respectively).The second alternative was to check the number of pests and spiders (18% vs 14%)early in the morning or late in the afternoon. Very few farmers applied knowledgegained about beneficial insects. Future training activities for farmers should focusmore on practical aspects of pest control.Table 4. Farmers’ information sources on farming practices.TopicAttendance at seminarFarmingRespondents who attended (n = 50)Pest managementIPM aRice productionCooperativeFarmers' Field SchoolChemical useNIA zonal programHome food processingTechnology/information disseminationRespondents (n = 65) who received informationSourceTopicRepresentative Chemical use(n = 38) Monitor pests-NEMaintain fieldNo responseDAMonitor pests-NE(n = 24) Chemical useNo responseNGO/radio Monitor pests-NEin = 3)Chemical useNo responseFarmers’ pest control practicesRespondents who did not attend (n = 43)Respondents who attended (n = 57)Observe pests and spiderslrrigateMaintain fieldOwn wayDo not knowNo responseIrrigate/sprayObserve pests and spidersIrrigateOwn wayNo responseResponse(%)282016166444235.44.61.516.921.59.26.21.51.51.5(%)14.09.32.32.32.348.822.817.515.81.842.1a IPM = Integrated pest management, NIA = National lrrigation Administration, DA = Department of Agriculture, NGO =nongovernment organization, NE = natural enemies.Community-based rice pest management 113

Table 5. Rice cultural management practices of 100 farmer-respondentsin Bual Norte.Planting Planting month/variety %TimeFirst crop (n = 107 a )Second crop (n = 107)Variety planted (n = 105 a )a Multiple answers.JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberIR62IR74IR56IR66IR8IR1314PSB-Rc 18PSB-Rc 10PSB-RC 34PSB-Rc 4PSB-RC 6PSB-RC 20Farmers' selection numberFarmers' selection series8.913.34.40.55.98.912.38.910.86.90.518.718.118.13.81.01.01.012.31.93.81.01.00.926.76.7Respondents’ rice culture management practices. During the first year of implementingthe NIA zonal irrigation program in 1996, staggered planting prevailed inBual Norte and Bual Sur (Table 5). To cope with the irrigation schedule, more farmersshifted from transplanting (TP) to direct seeding (DS). Since then direct seedinghas predominated (81%) in most villages of Midsayap. According to farmers, thelabor cost was lower in DS than in transplanting. Herbicides were commonly used forboth direct seeding and transplanting (1.0 L ha -1 ). DS was also easier and faster thanTP. Under normal weather conditions, most farmers practiced DS in both the dry andwet seasons to shorten crop maturity by about 10 d compared with TP, to get returnson investment more quickly, and to be able to pay back their loans as soon as possible.Some claimed that a crop established by DS had a low infestation of pests such asRBB, WSB, and RTD. The verified trials showed that only green leafhopper-resistantvarieties such as PSB Rc 34 established by DS and at a high seeding rate (120 kg seed114 Truong et al

ha -1 and above) had a lower RTD incidence than in TP. There was no significantdifference, however, in the cumulative disease incidence for GLH-susceptible IR64at 45 DAT, whether established by DS or TP.Several farmers used to plant two varieties or change varieties every season toevaluate varietal performance. They looked for resistant varieties to get good harvests.Seed exchange among farmers, friends, and relatives was a common practice.They tried three groups of varieties: those released by IRRI (45%), PhilippineSeedboard commercial rice varieties or PSB Rc (22%), and farmer selections (33%).Farmers preferred GLH-resistant varieties such as IR56 (7%), IR62 (18% ), IR74(18%),PSB Rc 10 (27%), PSB Rc 18 (12%), and PSB Rc 31 (4%) for rice tungro management.Information came from training or from friends who participated in the Farmers'Field School. Navarro et al (1998) cited a similar observation in a report on theoverview of the National Integrated Pest Management Program in the Philippines.Farmers who planted GLH-resistant varieties also applied insecticides at the rate of1.5 L ha -1 for a broadcast crop and 1.1 L ha -1 for a transplanted crop because theywanted 10 get a clean crop and avoid crop failure. This practice usually reduced thenatural enemies established early in the crop, resulting in pest infestation, whichprompted farmers to increase the insecticide application per season.All farmers in Bual Norte and some from Bual Sur villages practiced duck pasturingand golden snail culture successfully. Not all farmers raised ducks, but a fewhad large-scale duck farms for egg production, which contributed to the managementof golden snail. A farmer with 200 ducks could efficiently grow golden snails on a l-ha farm.Influence of sociodemographic, factors on respondents' cultural practices. Ricecropping practices of respondents at Bual Norte and Bual Sur villages can be categorizedinto four groups (Fig. 3). The first group (Fig. 3. top left corner) is composed ofyoung farmers (7%) who owned a farm larger than 1 ha. had less than 10 yr of farmingexperience, attained the highest education (2-yr college), and benefited the mostfrom technology or information dissemination and training courses. They usuallyplanted the IR varieties mentioned earlier. The second group (Fig. 3. bottom rightcorner) consists of tenant farmers (30%) who had less than 1 -ha farms, 21 to 30 yr offarming experience, planted farmers' selected varieties, and had RTD concerns, butnever received any technical information. The third group (top right corner) is thewomen farmers (16%) who had farms larger than 2 ha, finished high school, practicedsynchronous transplanting of PSB Rc varieties, and used to apply insecticidesagainst RBB infestation. Group 4 (bottom left corner) is the resource-poor farmers(41%) who only had elementary education, cultivated their land for more than 30 yr,practiced asynchronous planting and direct seeding, and were concerned about WSBdamage. The data showed that technology/information dissemination through trainingin the past had reached only a small proportion of educated and progressive farmers(group I). These farmers were tapped as information sources of IK while thecampaign paid more attention to the majority and resource-poor farmers (groups 2and 4). Group 4 farmers were requested to share their experiences on RBB controlpractices during the farmers' field day.Community-based rice pest management 115

Fig. 3. Correspondence chart showing the coherence of sociodemographic characteristics of farmerrespondentsand their rice cultural practices in response to rice black bug (RBB), white stem borer(WSB), and rice tungro disease (RTD) concerns, Bual Norte and Bual Sur, 1995-96: a guide toidentifying target farmer-participants.Monitoring of insects and diseases, and verified trialsThe weekly population of RBR, GLH, ZLH, WBPH, and WSB catches was monitoredby a light trap and sweeps from farmers' fields. The populations of GLH andZLH during 1997 were much higher than those in 1998 (Fig. 4, left and right). Hopper catches during 1997 from the light trap had three peaks, in February-March (A).June-July (B), and November-December (C). These population fluctuation patternsmirrored those recorded from farmers' fields, which had been established during peakperiods (D) and (F). Frequently, GLH and ZLH were caught together with MRB, andseparated from WSB and WBPH (C). The hopper population was lowest in farmers'fields planted during April-May (E). During the second peak of GLH and ZLH, therewas a risk of planting susceptible varieties IR60, PSB Rc4, Bugos, Selection #78.Series, and Masipag, which had high RTD incidence (30–80%). This finding stronglysuggests that farmers should replace these varieties with GHL-resistant ones duringthe second crop or in the wet season. Another alternative is the use of RTD virusresistantlines. Thirty farmer-cooperators in Bual Norte successfully tested IR71031-4-5-5-1 for RTD resistance for the first time in the 1998 wet season. This RTD-resistantadvanced line yielded 3.5-4.5 t ha -1 of rice.116 Truong et al

Fig. 4. Three major peaks of green leafhopper (GLH). zigzag leafhopper (ZLH), whitebacked planthopper(WBPH), and mirid bug (MRB) population from light trap (A, B, C) in relation to population onfarmers’ fields at Bual villages in 1997 (D, E, & F), and tree diagram of their population linkage (G),and weekly population patterns in 1998 (H).Likewise, RBB catches were also extremely high in 1997 (Fig. 5), with two peaks,in February-March and October-November (A). In a verified trial established duringthe 1997 dry season, only the stopgap line IR1314, variety C4-137, and a few farmerselections yielded from 1 to 2.5 t ha -1 (B). The field population of RBB reached 10adults hill and induced 30-45% RBB dead hearts (C). Similar entries planted duringthe wet season suffered a higher incidence of RBB deadhearts (30–60%), and RBBwhiteheads (30–80%), and RBB burns (90-100%) (D). In contrast the crop grown inBual Norte from the end of March to mid-July in 1998 after a long fallow period(January-March) had a negligible RBB population and RTD incidence (data not shown).Community-based rice pest management 117

Fig. 5. Monthly population pattern of RBB catches in the light trap (A) and high bug infestationlevels in the transplanting trials during dry (B) and wet (C) seasons. Only few varieties were tolerantof RBB (D) (Midsayap Experiments Station, 1997).Evaluation of community cultural practices and crop performanceThree farmer communties participated in rice pest management activities for fourseasons (Bual Norte, 1996-98), three seasons (Bual Sur, 1996-98), and two seasons(Bobonao, 1997-98). A combined data analysis showed that, in all communites theuse of insecticides was possitively correlated with whiteheads (%) caused by WSB(Figs. 6 and 7). This implies that farmer communities did not benefit from insecticidesused against WSB. Their grain yields (3.2–4.6 t ha -1 ) were not affected bywhiteheads below 20%. Grain yields started to decline only when whitehead incidencesurpassed 20% (data not shown). Future farmer forums should thus emphasizethe use of natural enemies in the management of WSB, particularly during early infestationat the vegetative growth stage, to reduce the insecticide application frequency.In Bual Norte, a combination of varieties and direct seeding with no insecticide usewas an effective control component (7% of farmers) in RTD management, with 2118 Truong et al

RBB hill -1 and 4% WSB (Fig. 6). In Bual Sur, direct seedings effectively reduced theRBB population (1.3 hil -1 ; data not shown), but RTD remained a problem in the July1997 planting (12% of farms). Results indicated that a community strategy in RTDmanagement should focus on varietal deployment and a change in planting schedule.Fig. 6. Principal component analysis of rice cultural practices of farmers in Bual Norte village inrelation to their pest control measures from 1996 to 1998 showed that grain yields (not frequencyof insecticides used) were positively correlated with a combination of direct seeding and varietiesagainst rice tungro incidence and black bug density under a 56-d planting period.Fig. 7. Varieties planted and frequency of insecticides used by farmers in Bobonao village (notplanting period or method) effectively protected grain yield from damage caused by rice black in1997. Planting method, however, correlated positively with rice tungro incidence, which was lowduring the same period and did not affect overall the harvest of the community.Community based rice pest management 119

In Bobonao, RTD (2% of farms) did not adversely affect community rice production(4.6 t ha -1 ), as only a few fields were infected with RTD (Fig. 7). Planting dates werepositively correlated with the RBB population, which was effectively controlled by acombination of varieties and insecticides used.Conclusions1. The project used the PRA method in the baseline survey through the participationof farmer communities at Midsayap, North Cotabato, and collaborative elfortsof various agencies concerned about rice production and pest problems todevelop a conceptual framework.2. An informal group-focused dialogue was the key strategy in the PRA methodused to discover farmers’ IK on rice cultural practices and to design the actionplan on community pest management for 1996-98.3. Little progress has been made in the past 2 yr to enrich farmers’ IK on crop-pestmanagment by shortening the period of planting in each village, to shift fromtransplanting to direct seeding, and to reduce the frequency of insecticide application.The introduction of RTD-resistant lines has just started. The use of naturalenemies in crop management in farmer communities has not been widelyadopted.4. Understanding the rice ecosystem through continuous learning would enhancecommunity cooperation, which plays a vital role in managing pests and sustainingrice production.ReferencesJusto HD. 1995. Integrated management of the Malayan black bug. Rice Tech. Bull. No. I.DA-PhilRice, Maligaya, Muñoz, Nueva Ecija. 12 p.Navarro RL. Medina JR. Callo DP. Jr. 1998. Empowering farmers: The Philippine NationalIntegrated Pest Management Program. SEAMEO Regional Center for Graduate Studyand Research in Agriculture. Technical Bulletin. Los Baños. Laguna: SEARCA. 114 p.Sanchez LN. Obien SR. 1995. Profile of insect pests and diseases in Mindanao. In: The 3rdNational Rice R & D Review and Planning Workshop, 1–3 March 1995. PhilRice. p 21–36.Tiongco ER. Flores ZM. Lapis DB. 1996. Rice tungro virus disease. Rice Tech. Bull. No. 15.DA-PhilRice. Maligaya. Muñoz. Nueva Ecija. 76 p.NotesAuthors’ Address: X.H. Truong, E.H. Batay-an. S.C. Mancao. G.N.A. Abrigo. A.B. Estoy, L.B.Flor, Jr. H.D. Justo, Jr, E.R. Tiongco, R.N. Casco, and S.R. Obien, Philippine Rice ResearchInstitute, Maligaya, Muñoz 3119 Nueva Ecija, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL. editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998. IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.120 Truong et al

The influence of varietal resistance andsynchrony on tungro incidence in irrigatedrice ecosystems in the PhilippinesT.C.B. Chancellor, E.R. Tiongco, J. Holt, S. Villareal, and P.S. TengRice tungro disease is endemic in some intensively cultivated areas in thePhilippines where planting dates are highly asynchronous. Results from asurvey conducted in Albay Province in 1992-94 showed that tungro diseaseincidence was significantly lower on rice varieties that are resistantto the main leafhopper vector of tungro, Nephotettix virescens, than onsusceptible varieties. Late-planted rice crops in both the wet and dry seasonshad the highest incidence of tungro. In on-farm trials in North CotabatoProvince, a virus-resistant advanced breeding line, lR68705-1-1-3-2-1,showed strong resistance to tungro. These findings are discussed in relationto developing optimal tungro management strategies for endemic areas.Rice tungro disease was recorded as early as 1859 in Indonesia (Ou 1985) but it onlybecame a serious problem in South and Southeast Asia in the 1960s as a result ofmajor changes in rice production systems during that decade. The introduction ofmodern, photoperiod-insensitive, semidwarf indica varieties with a good response tofertilizer led to substantial increases in productivity (IRRI 1985). Many of these varietieshad significantly shorter maturity periods than traditional varieties. This allowedan increase in cropping intensity, so that two or even three rice crops could be grownin a year. Large yield increases were also due to an expansion in area planted to riceand heavy investment in irrigation schemes that increased water use efficiency andmade intensive production possible.Although changes in rice cultivation practices increased grain yields, the intensificationof rice production was associated with more severe insect pest and diseaseproblems. Multiple rice cropping over large areas provided a continuous source ofplant hosts and enabled the year-round development of insect pests (Litsinger 1989).For rice virus diseases, such as tungro, the potential survival of pathogens was enhancedby the greater continuity of hosts (Thresh 1989). This was particularly importantin areas with highly asynchronous planting dates and overlapping rice crops.The frequency of major tungro outbreaks appears to have decreased since theypeaked in the late 1960s and early 1970s, although large areas of rice were affected bytungro in some eastern coastal areas of India in 1990 and in central Java, Indonesia, in1995. In the Philippines, the last major tungro epidemic occurred in 1983-84 (Baria1997), although the disease is present every year in some endemic areas (Savary et al1993). In a survey of historical data from 1989 to 1993 on insect pest and diseaseincidence in Mindanao, tungro was considered to be the most destructive disease ofrice (Sanchez and Obien 1995). Recent survey data confirm that tungro is a continuingproblem in Mindanao (Truong et al, this volume). In this chapter, we discuss therole of two of the most important factors that affect tungro incidence in irrigated riceecosystems in the Philippines—varietal resistance and synchrony of planting dates.

Varietal resistanceFew commercial varieties with virus resistance are grown in the Philippines, althoughadvanced breeding lines with resistance to tungro viruses are currently being developedat IRRI (Cabunayan et al, this volume). Several different sources of resistanceto Nephotettix virescens (Distant), the main leafhopper vector, however, have beenused in breeding programs and deployed over large areas (Khush 1984). The use ofleafhopper-resistant varieties has been one of the major strategies for tungro managementand has been highly effective. Rice varieties with resistance to N. virescens arenot readily infected with tungro viruses except where disease pressure is high. Resultsfrom a survey conducted in the municipality of Polangui, Albay Province, Luzon,indicated that farmers clearly benefited from growing leafhopper-resistant varietiesbecause of reduced tungro incidence (Chancellor et al 1996). Nevertheless, adaptationof leafhopper populations to previously resistant varieties has occurred in someareas, particularly in certain provinces in Mindanao (Dahal et a1 1990).In the Polangui survey, adaptation of leafhopper populations is not related toincreased leafhopper number. Although the field resistance to tungro of resistant varietieswas due to leafhopper resistance, leathopper number was similar on resistantand susceptible varieties. Figure 1 shows the mean peak numbers of adults and nymphsof N. virescens collected by 10 sweeps of a 30-cm-diameter insect net in fields with atleast 1% tungro-affected rice hills. The mean peak number per 10 sweeps of N. virescensadults was 35 for both resistant and susceptible varieties. Nymphal counts were alsosimilar at 30 and 25 for resistant and susceptible varieties, respectively. Leafhopperstend to feed in the xylem, rather than the phloem, of resistant varieties (Auclair et al1982) and this may account for the lower incidence of tungro observed on resistantvarieties in Polangui. For all these reasons, it is important that the reaction of resistantvarieties to tungro disease be closely monitored after resistance deployment in differentrice-growing areas and that information be disseminated to farmers through extensionagencies.Fig. 1. Tungro disease incidence and abundance of adults and nymphs, respectively, of Nephotettixvirescens on leafhopper-resistant and -susceptible rice varieties in farmers' fields in Albay Province,Luzon, Philippines, planted between November 1992 and October 1994. Bars indicate 95% confidencelimits.122 Chancellor et al

Limited data exist on how resistant varieties perform when grown widely underfield conditions. Varieties IR60, IR62, IR66, IR70, IR72, and IR74, which are resistantto N. virescens but not to tungro viruses, were classified as “resistant” based ontheir field reaction to tungro in multilocational varieta1 trials conducted by PhilRice.All other varieties grown in the study area were classified as susceptible. Figure 1shows that in fields planted between November 1992 and October 1994 in which atleast 1% rice hills were affected by tungro, mean peak disease incidence on resistantvarieties was 4.5% compared with 11.5% for susceptible varieties.On-farm varietal trials were conducted in North Cotabato in 1996 using a methodologysimilar to that described in Subramanian et al (this volume). Tungro incidencewas low in two of the trials but was high in San Pedro, where the highly susceptiblevariety Masipag, which was chosen by the farmer. was severely affected bytungro at 42 d after transplanting (DAT) and yielded only 0.5 t ha -1 (Table 1). Incontrast. green leafhopper-resistant IR62 and virus-resistant IR69705-1-1-3-7-1 hada much lower tungro incidence in spite of the strong sources of inoculum in Masipagplots and surrounding areas. In IR69705-1-1-3-7-1, the result is particularly strikingas leafhopper numbers were very high but only 2% of the hills were affected at 56DAT. The lower than expected yields in IR62 and IR69705-1-1-3-2-1 plots were mainlyattributed to damage caused by a late attack of the rice black bug, Scotinopharacoarctata.Synchrony of planting datesLoevinsohn (1984) reported that in the 1981 wet season in an irrigated, double-croppedrice area in Nueva Ecija, Luzon, tungro prevalence in six sites increased with asynchronousplanting within 0.6 km of the sites. Sanchez and Obien (1995) consideredasynchronous planting dates to be one of the major factors affecting tungro preva-Table 1. Tungro disease incidence a , green leafhopper (GLH) numbers b , and yield in different lines andvarieties in on-farm trials in the barangays of Central Bulanan, San Pedro, and Villarica, Midsayap,North Cotabato, 1996 wet season.Site Line/variety GLH no. Tungro Yield (t ha -1 )Incidence (%)Central BulananSan PedroVillaricaIR62lR68305-18-1PSBRc10IR62lR69705-1-1-3-2-1MasipagIR62lR68305-18-1Line 60164.0 ± 7.5318.7 ± 77.3287.8 ± 18.733.7 ± 2.2230.5 ± 28.9494.8 ± 41.725.1 ± 2.1183.0 ± 3.568.9 ± 18.42.4 ± 0.74.7 ± 0.94.1 ± 0.916.3 ± 5.73.7 ± 0.495.0 ± 5.09.6 ± 3.218.7 ± 2.20.5 ± 0.53.2 ± 0.52.6 ± 0.22.9 ± 0.21.5 ± 0.21.3 ± 0.10.5 ± 0.13.0 ± 0.12.1 ± 0.13.5 ± 0.1a Number of tungro-diseased hills at 56 d after transplanting (DAT) or seeding (DAS). Mean of three replications.b Average numbers of adults and nymphs of Nephotettix virescens per 10 sweeps of a 30-cm-diameter insect netcollected over three sampling dates at 28, 42, and 56 DAT or DAS (42 and 56 DAT only for Central Bulanan). Mean ofthree replications.The influence of varietal resistance and synchrony 123

lence in Mindanao from 1989 to 1993. Continuous planting with no fallow periodbetween rice crops and wide variation in planting dates were regarded as the maincauses of tungro problems in Camarines Sur and Albay Provinces (Baria 1997).Figure 2 shows the distribution of planting dates for rice fields from November1992 to October 1994 in the survey area in Polangui. Planting was carried out monthlyfrom November 1992 to March 1994. Although planting was continuous, peaks wereobserved in the number of fields planted within this period, which corresponded toeach of the three main cropping seasons. These peaks occurred in December andJanuary (1992-93 dry season), May to July (1993 wet season), and January and February(1993-94 dry season). No clear break between cropping seasons, however, wasobserved during this period. In contrast, there was a distinct gap between the 1993-94dry-season and the 1994 wet-season crops, as only one field was planted in April andnone in May. Most of the 1994 wet-season plantings were carried out in June and Julywhen few standing rice crops remained in the area.Figure 2 also shows the mean peak tungro incidence for all fields recorded in thesurvey for each month of planting from November 1992 to October 1994. In both the1992-93 and 1993-94 dry seasons, tungro disease incidence was greatest in Marchplantings. In the 1993 wet season, tungro incidence was greatest in August and Sep-Fig. 2. Frequency of planting of rice crops and mean peak tungro disease incidence in farmers’ fieldsin Albay Province, Luzon, Philippines, planted between November 1992 and October 1994.124 Chancellor et al

tember plantings, reaching 13% and 14%, respectively. Thus, in each of these seasons,late plantings were seriously affected by tungro. Staggered planting allowed theinoculum to be carried over from earlier to later plantings within a season. Diseasedeveloped in some of the fields planted during the main planting periods, but incidencerarely reached high levels. This disease increase, however, was sufficient toresult in significantly high inoculum levels at a time when later plantings were at avulnerable growth stage.Tungro disease incidence was lower in the 1994 wet season than in previousseasons. The fallow period following the 1993-94 dry season reduced the inoculumso that there was less risk of infection for wet-season plantings. Furthermore. most ofthe 1994 wet-season rice crops were planted in June and July, with few higher risklate-season plantings. Although the effects of environmental factors on tungro incidencewere not noted during the survey, the 2-mo break in planting appeared to havehelped reduce tungro incidence in the 1994 wet season.Implications for tungro management strategiesIn the 1992 wet season, 340 ha of rice were sprayed with insecticide in Polangui,Albay Province, in a coordinated campaign conducted by the Department of Agricultureto contain what was thought to be a developing outbreak of tungro disease. Sprayinginsecticide against leafhoppers in response to tungro remains a standard practicein many areas in the Philippines, even though the effectiveness of this approach maybe uncertain. In interviews conducted with rice farmers in five provinces in the Philippinesin 1994, many respondents thought that insecticide application did not controlthe spread of tungro very effectively (Warburton et al 1997). The continued practiceof spraying is due in part to the perceived advantage of protecting the crop againstother insect pests.Many potentially adverse side effects of spraying insecticides in rice have beenwell documented. Because of the limited efficacy of spraying insecticides againstleafhoppers for tungro control, and the ineffectiveness of other tactical measures suchas roguing (Tiongco et al 1998), strategic measures appear to offer the best prospectsfor tungro management in the Philippines (Holt et al 1996). In South Sulawesi, Indonesia,a tungro management scheme was introduced in the mid-1980s in which plantingdates were synchronized and targeted to avoid periods of peak vector populations(Sama et al 1991 ). The scheme also involved varietal rotation to reduce the potentialfor leafhopper adaptation to resistant varieties, and selective recommendations forvector control using insecticides. Tungro incidence has remained low since the introductionof the scheme, except in small areas where it has been difficult to synchronizeplanting dates (S. Sama, personal communication, 1994). The scheme has been quiteeffective and synchronous planting may have played a key role in its success.Results from the survey conducted in Polangui are also consistent with the notionthat planting rice crops synchronously reduces the risk of tungro incidence considerably.A mathematical model was developed and used to examine the area-widedynamics of tungro disease (Holt and Chancellor 1997). The potential impact ofThe influence of varietal resistance and synchrony 125

changes in cropping synchrony in reducing tungro disease levels was assessed. Modeloutputs showed that tungro endemicity was determined mainly by planting date varianceand disease persisted if this variance exceeded a certain threshold. Where plantingdates were only moderately asynchronous, relatively small reductions in plantingdate variance made a significant contribution to reducing tungro incidence. If theplanting was highly asynchronous, however, then a similar marginal increase in synchronyonly slightly reduced tungro incidence. Consequently, where rice croppingpatterns are highly asynchronous and it is impractical to change them dramatically, adifferent management approach is needed. In such situations, varietal resistance offersthe best prospects for success.Varietal selection by farmers is often conditioned more by factors such as eatingquality and market price of the grain than by pest and disease resistance characteristics.In areas with high tungro incidence, however, resistance characteristics are animportant criterion for varietal selection (Warburton et al 1997, Warburton et al, thisvolume). The mathematical model of tungro dynamics was used to examine issuesrelated to the deployment of resistant varieties (Holt and Chancellor, in press). Themost effective strategy, based on model outputs, was to concentrate deployment ofresistant varieties in the wet season, the season of greatest disease spread. A relativelylarge proportion of fields need to be planted with resistant varieties to significantlyreduce tungro incidence in fields planted with susceptible varieties. Nevertheless,results from the on-farm trial in San Pedro and the data from many seasons of onstationtrials in Midsayap and Maligaya (Cabunagan et al, this volume) show theadvantages of growing resistant varieties. Therefore, the deployment of resistant varietiesmay considerably benefit farming communities, even if the very high levels ofadoption possible in South Sulawesi are not obtained.ReferencesAuclair JL, Baldos E, Heinrichs EA. 1982. Biochemical evidence for the feeding sites of theleathopper Nephotettix virescens within susceptible and mistant rice plants. Insect Sci.Applic. 3:29–34.Baria AR. 1997. Status of rice tungro disease in the Philippines: a guide to current and futureresearch. In: Chancellor TCB, Thresh JM, editors. Epidemiology and management of ricetungro disease. Chatham (UK): Natural Resources Institute.Chancellor TCB, Tiongco ER, Holt J, Villareal S, Teng PS, Fabellar N, Magbanua MGM.1996. Risk factors for rice tungro disease in endemic areas. In: Chancellor TCB, Teng PS,Heong KL, editors. Rice tungro disease epidemiology and vector ecology. IRRI Discuss.Pap. Ser. No. 19.Dahal G, Hibino H, Cabunagan RC, Tiongco ER, Flores ZM, Aguiero VM. 1990. Changes incultivar reactions to tungro due to changes in “virulence” of the leafhopper vector. Phytopathology80:659–665.Holt J, Chancellor TCB. 1997. A model of plant virus disease epidemics in asynchronouslyplanted cropping systems. Plant Pathol. 46:490–501.Holt J, Chancellor TCB. Modelling the spatio-temporal deployment of resistant varieties toreduce the incidence of rice tungro disease in a dynamic cropping system. Plant Pathol.(in press).126 Chancellor et al

Holt J, Chancellor TCB, Reynolds DR, Tiongco ER, 1996. Risk assessment for rice planthopperand tungro disease outbreaks. Crop Prot. 35:359–368.IRRI. 1985. International rice research: 25 years of partnership. Los Baños (Philippines): InternationalRice Research Institute.Khush GS. 1984. Breeding rice for resistance to insects. Prot. Ecol. 7:147–165.Litsinger JA. 1989. Second generation insect pest problems on high-yielding rices. Trop. PestManage. 35:235–242.Loevinsohn ME. 1984. The ecology and control of rice pests in relation to the intensity andsynchrony of cultivation. Ph.D. thesis. University of London, London.Ou SH. 1985. Rice diseases. Wallingford: Commonwealth Agricultural Bureau.Sama S. Hasanuddin A, Manwan I, Cabunagan RC, Hibino H. 1991. Integrated management ofrice tungro disease in South Sulawesi, Indonesia. Crop Prot. 10:34–40.Sancher LM, Obien SR. 1995. Profile of insect pests and diseases in Mindanao. Paper presentedat the 8th National Research and Development Review and Planning Workshop,1–3 March 1995, PhilRice, Maligaya, Muñoz, Nueva Ecija.Savary S. Fabellar N, Tiongco ER, Tens PS. 1993. A characterization of rice tungro epidemicsin the Philippines from historical survey data. Plant Dis. 77:376–382.Thresh JM. 1989. Insect-borne viruses of rice and the Green Revolution. Trop. Pest Manage.35:261–272.Tiongco ER, Chancellor TCB, Villareal S, Magbanua M, Teng PS. 1998. Roguing as a tacticalcontrol for rice tungro virus disease. J. Plant Prot. Trop. 11:45–57.Warburton H, Palis FL, Villareal S. 1997. Farmers’ perceptions of rice tungro disease in thePhilippines. In: Heong KL, Escalada MM, editors. Pest management practices of ricefarmers in Asia. Manila (Philippines): International Rice Research Institute. p 129–141.NotesAuthors' addresses: T.C.B. Chancellor and J. Holt, Natural Resources Institute. University ofGreenwich. Central Avenue, Chatham Maritime, Chatham, Kent ME4 4TB, UK; E.R.Tiongco, Philippine Rice Research Institute, Maligaya, Muñoz, Nueva Ecija 3119;S. Villareal and P.S. Teng, International Rice Research Institute. MCPO Box 3127, MakatiCity 1271, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.The Influence of varietal resistance and synchrony 127

Improving IPM technology for rice tungrodisease in IndonesiaA. Hasanuddin, I.N. Widiarta, and YuliantoRice tungro disease (RTD) is of great economic importance in Indonesia.RTD distribution is still expanding and the disease causes Serious outbreaksin some seasons. In the 1994-95 wet season, RTD severely attackedrice plants in East Java and in the Surakarta region of Central Java.The yield loss was estimated to be approximately 25 billion Indonesianrupiah. RTD is successfully controlled in South Sulawesi by integrated pestmanagement (IPM), combining planting at the appropriate time and theuse of green leafhopper-resistant varieties in rotation. This control methodneeds to be adapted, however, before it can be applied elsewhere, especiallyin areas planted asynchronously. Some experimental activities havebeen carried out to improve IPM for tungro such as determining the minimumarea necessary to conduct synchronized planting, adjusting plantingtime, adapting Nephotettix virescens to resistant varieties as the basis ofvariety rotation, and using selective weed sanitation to reduce infectionsources. Experimental results indicated that tungro disease spread from asingle source of diseased plants reached 200 m. In asynchronously plantedfields, RTD incidence at harvest in an observed field correlated positivelywith RTD incidence at 6-10 wk after transplanting (WAT) in an area within aradius of 101–250 m when rice plants in the observed field were at 3 WAT.RTD incidence was high regardless of planting time in asynchronouslyplanted fields. In synchronously planted fields, however, the later the plantingtime, the greater the disease incidence. RTD incidence in Bali, West Java,and Central Java was high on Cisadane and IR64, which possess the Glh5resistance gene. N. virescens colonies collected from West Java, CentralJava, Bali, and South Sulawesi were well adapted to IR64. Transmission ofrice tungro viruses was achieved in six weed species. This information canbe used to improve IPM for RTD in Indonesia.Rice tungro disease (RTD) is efficiently spread by green leafhopper (GLH) species,especially Nephotettix virescens Distant. In the 1994-95 crop season, an outbreak ofRTD occurred in East and Central Java. The yield loss approached 25 billion Indonesianrupiah (Anonymous 1995). The area affected by RTD is still expanding, especiallyin West Java, where RTD was found in the mountainous area but not in thenorthern coastal lowland until the 1996-97 wet season (Hasanuddin et al 1995, Widiartaet al 1997). In the 1996-97 wet season, RTD was found in Sukamandi (Widiarta et alunpublished). RTD has the potential to cause serious outbreaks.Sama et al (1991) reported that RTD was controlled successfully in South Sulawesiby integrated pest management practices, combining planting at the appropriate timeand the use of GLH-resistant varieties in rotation. Recommended transplanting dateswere based on the seasonal fluctuation of rainfall, tungro incidence, and green leafhopperpopulations. These dates were selected to avoid periods of high disease pressure.Varieties resistant to N. virescens are categorized into five groups, based onresistance genes present, for variety rotation. T0 varieties have no resistance gene.The other groups—T1, T2, T3, and T4—have resistance genes Glh1, Glh6, Glh5, and

glh4, respectively. Planting at the appropriate time was considered the most importantfactor in avoiding periods of high disease pressure. Before the implementation ofIPM, the affected area exceeded 1,000 ha, but recently this has dropped to only 100ha. Implementing IPM for tungro in provinces outside South Sulawesi is more difficult,especially in the asynchronously planted areas of Java and Bali. Consequently,RTD is most prevalent in these islands, which contribute more than 60% of the totalrice production in Indonesia. An RTD management strategy suitable for asynchronouslyplanted areas of Java and Bali is urgently needed to minimize yield loss.Experiments were conducted to estimate the minimum area necessary to undertakesynchronized planting, adjust planting dates, adapt N. virescens to resistant varietiesas the basis of variety rotation, and use selective sanitation to reduce sources ofRTD inoculum. The objective of these experiments was to improve IPM implementationin tungro-endemic areas in Java and Bali. The results are reported here.Materials and methodsUnit of synchronized plantingSynchronized planting within a minimum area is necessary to benefit from implementingrecommended transplanting times. The size required was estimated by twofield experiments. First, the disease gradient from a single virus source was recorded.Second, RTD incidence in an observation field was correlated with RTD incidence inthe surrounding area.Disease gradient distribution. This experiment was conducted in Sidrap in the1996 dry season and in Maros in the 1996-97 wet season in a 10-ha rice area that wasplanted synchronously. Preinfected seedlings of Cisadane, the inoculum source, wereplanted at a spacing of 20 × 20 cm in the center of the study area in a 10 x 10-m plot.Monitoring plots were established at distances of 100, 300, 300, 400, and 500 m fromthe inoculum source and were planted with disease-free seedlings of Cisadane at 20 ×20-cm spacing in 10 × 10-m plots. The green leafhopper population was observed at6 WAT by 10 sweeps of an insect net and RTD incidence was assessed at 8 WAT in themonitoring plots.Relationship between tungro incidence and diseases in observed fields. This fieldexperiment was conducted in an asynchronously planted area at Subak Padang Galak,Badung regency, Bali, during the 1995 dry season and 1995-96 wet season in a 12-harice field. Rice fields in the area were mapped, and rice plants were planted monthlyduring the wet and dry seasons for observation. The population densities of greenleafhopper and tungro incidence were assessed when rice plants in the observed fieldwere at 3, 8, and 12 WAT. At the same time, GLH population density and tungroincidence were observed in surrounding fields 50, 100, 200, and 250 m away. Ricestages in the area were also mapped and categorized into seedling, young rice plant(1–5 WAT), older rice plant (6–10 WAT), and stubble for all surveys.130 Hasanuddin et al

Time of planting recommendationVarieties in each of the five group (T0 to T4) were planted at early, normal, and lateplanting times in the synchronously and asynchronously planted fields. Early plantingwas at the same time as the earliest plantings by farmers. Normal and late plantingwere 1 and 2 mo later, respectively. The experiment was conducted in Bali, CentralJava, and West Java in the 1996 dry season and 1996-97 wet season. GLH populationdensity and tungro incidence were observed at 4, 6, and 8 WAT. The population densityof green leafhoppers was surveyed by 20 strokes of an insect net and RTD incidencewas assessed from 100 hills selected randomly.Variety rotationPopulations of N. virescens were collected from rice fields in West and Central Java,Bali, and South Sulawesi from the dominant variety in the particular area. Each colonywas reared separately in the greenhouse. Adults from each colony were placed ontungro-infected rice plants in an insect cage for 4 d. Each colony was allowed totransmit virus on a set group of varieties with different genes for resistance to GLH(T0 to T4) for 1 d. The percentage of tungro-infected plants was observed 2 wk afterinoculation feeding. The level of resistance to green leafhopper as indicated by thesurvival rate of colonies was tested by introducing five second or third instar nymphsfrom each colony into test tubes with 14-d-old seedlings of a set group of varietieswith T0-T4 resistance genes. Survival rates and duration until the adult stage wereobserved.Selective sanitationThis study involved a two-step experiment. First, weed species that could be successfullyinoculated with rice tungro viruses by N. virescens were determined. Second, atest was conducted to see whether N. virescens could obtain virus from those weedsand successfully transfer the virus to rice plants.In the first experiment, weeds commonly found in rice fields were collected andtransplanted individually in pots. The weed plants were collected from epidemic andendemic areas of RTD in Bali, Klaten. and Yogyakarta. Each weed species was inoculatedseparately by exposing each plant for 1 d to four infective N. virescens that hadfed for 4 d on rice plants infected with rice tungro bacilliform (RTBV) and rice tungrospherical (RTSV) viruses. In the second experiment, seedlings of rice variety IR64were inoculated by N. virescens that had fed on weed species infected with RTBV andRTSV. At 35 d after inoculation, extracts of weeds and rice seedlings were indexed byenzyme-linked immunosorbent assay (ELISA) for the presence of RTBV and RTSV.lmproving IPM technology 131

Results and discussionMinimum unit for synchronized plantingRTD incidence in the inoculum source field was 89%. RTD incidence was low in allplantings away from the infection source (Fig. 1). The population density of N. virescensfluctuated from 12 to 17 per 10 strokes.RTD incidence at harvest correlated positively with RTD incidence in an areawithin a radius of 101–250 m at 6–10 WAT, when rice plants in the observation fieldwere at 3 WAT (Fig. 2). Therefore, the area inside this radius can be considered as theunit that should be planted synchronously. Assuming that RTD spread equally in alldirections, the size of the synchronously planted area should be about 20–40 ha.Loevinsohn and Alviola (1991) reported a significant correlation between the degreeof asynchrony and tungro spread to plantings within a radius of less than 600 m.Planting time recommendationIn the asynchronous area, early and late-planted rice had a similar risk of infection byRTD (Fig. 3). In the synchronous planting, early planting showed the lowest risk ofRTD infection. Chancellor et al (1996) reported a similar finding in the Philippines.Savary et al (1993), however, found that low RTD incidence was associated with lateplanting rather than early and mean planting dates in the Philippines. The differencein results may be due to the different methods of assessing planting time. The resistancelevel of the variety influences RTD incidence in both synchronously and asynchronouslyplanted areas. The resistant variety in group T3 showed tungro incidencesimilar to that in the susceptible T0 group.Fig. 1. Incidence of tungro disease and population of green leafhopper (GLH) at different distancesfrom an inoculum source. DS = dry season, WS = wet season.132 Hasanuddin et al

Fig. 2. Correlation between tungro incidence in observation fields at 8 and 12 wk after transplanting(WAT), and incidence within a radius of 101-250 m in the surrounding area when rice plants in theobservation fields were at 3 WAT.Fig. 3. Tungro incidence in various groups of resistant varieties in synchronously and asynchronouslyplanted fields with rice plants planted in the early, middle, and late crop season. DS = dry season,WS = Wet season, MKL = Mungkul village, PDG = Padanggalak village, JKL = Jaten village.Improving IPM technology 133

Fig. 4. Transmission efficiency of colonies of Nephotettix virescens on rice varieties in five resistancegroups. Bars within a variety and colony with the same letter are not significantly different at the 5%level by Duncan’s multiple range test (DMRT).Variety rotationFigure 4 indicates that some of the resistant varieties tested have now become susceptible.N. virescens colonies tested on a set of varieties with different genes for resistanceshowed variations in their virulence. IR72 was not preferred by N. virescenscolonies from Bali or Central Java. The survival rate of N. virescens nymphs fromBali on IR72 was less than 10%, whereas the survival rate of West Java and SouthSulawesi colonies on IR72 was not significantly different from that of the controlvariety, Cisadane (Fig. 5). IR26 (Tl) and Ciliwung (T2) were still resistant to thecolony from South Sulawesi. Information on transmission ability and survival ratebased on the local situation can be used in choosing suitable varieties for varietalrotation.Although all N. virescens colonies used in this experiment were collected fromIR64, the degree of adaptation to the varietal resistance groups was variable. Thevariety composition in the area probably influenced the variability in the adaptationof N. virescens. Bastian et al (1995) showed that the survival rate of N. virescenscolonies differed according to the variety from which they were collected.134 Hasanuddin et al

Fig. 5. Survival rate of Nephotettix virescens reared on rice varieties from different resistance groups.Bars within a variety and colony with the same letter are not significantly different at the 5% level byDuncan’s multiple range test (DMRT).Selective sanitationAmong weed species inoculated by infective N. virescens, only Jussiaea repens,Trimthema portulacastrum, Phylanthus niruri, Cyperus rotundus, Monochoriavaginalis, and Leersia hexandra gave positive ELISA readings, suggesting that theseweeds can be infected by RTSV and RTBV (Table 1).Of the five successfully infected weeds ( L. hexandra, C. rotundus, M. vaginalis,J. repens, and C. difformis ), only C. rotundus and M. vaginalis served as an infectionsource of tungro viruses for rice.Rice tungro viruses reportedly survive in rice stubble, weeds. and wild rice species.which act as sources for reinfection of succeeding rice crops (Tiongco et al1992). Anjaneyulu et al (1988) reported that, of some weeds infected with RTSV,only Brachiaria mutica and Axonopus compressus were infected with RTBV.ConclusionsThe minimum area for synchronized planting is 20–40 ha. Planting time recommendationsare difficult to implement in asynchronously planted areas. Rotation of GLHresistantvarieties can be implemented in both synchronously and asynchronouslyImproving IPM technology 135

Table 1. Enzyme-linked immunosorbent assay (ELISA) reactionsof sap from plants inoculated with rice tungro spherical (RTSV)and bacilliform (RTBV) viruses by Nephotettix virescens.WeedEchinochloa colonum (L.) Link.E. crus-gall; (L.) DeauvLeersia hexandra SwartzLeptochloa chinensis (L.) Nees.Panicum repens L.Cyperus difformis L.C. rotundusC. halpan L.C. iria L.Jussiaea repensNeocharis pellucida Pres.Fimbristylis littoralis GaudichScirpus juncoides Roxb.Marsilea crenata Presl.Trianthema portulacastrumAlternathera philoxeroides Mart.A. sessilis (L.) D.C.Limnocharis flava (L.) BuchCommelina nudiflora L.Eclipta prostrata L.Ludwigia adscendens L.L. octovalvis (Jacq.) RavenMonochoria vaginalis (Burm.) Presl.Phylanthus nirurilmperata cylindricaELISA test reactionRTSV RTBV- -- -+++++++- -- -+- -- -+----- -+- ------ -- -- -- -++- -planted areas as long as N. virescens has not yet adapted to the varieties. Two weeds,Cyperus rotundus and Monochoria vaginalis, as well as ratoon crops should be eradicatedbefore nursery bed preparation. N. virescens is already adapted to IR64, but theother varietal resistance groups can still be used.ReferencesAnjaneyulu A, Daquioag RD, Mesina ME, Hibino H, Lubigan RT, Moody K. 1988. Host plantsof rice tungro (RTV) associated viruses. Int. Rice Res. Newsl. 13(4):30–31.Anonymous. 1995. Tungro outbreak in Central Java. Central Research Institute for Food Crops.15 p.Bastian AH, Talanca, Hasanuddin A. 1995. Transmission ability of some colony of green leafhopper(Nephotettix virescens) from the fields. A paper presented at the Annual Meetingof the Indonesian Society of Phytopathology, 6–8 September 1993, Yogyakarta, Indonesia.Chancellor TCB, Teng PS, Heong KL. 1996. Rice tungro disease epidemiology and vectorecology. Manila (Philippines): International Rice Research Institute. 104 p.Hasanuddin A, Widiarta IN, Yulianto. 1995. Status of rice tungro disease in Central Java andWest Java. A paper presented at the Annual Meeting of the Indonesian Society of Phytopathology,25–27 September 1995, Mataram, Indonesia.136 Hasanuddin et al

Loevinsohn ME, Alviola AA. 1991. Effect of asynchronized rice planting on vector abundanceand tungro (RTD) infection. Int. Rice Res. Newsl. 16:20–21.Sama S, Hasanuddin A, Manwan I. Cabunagan RC, Hibino H. 1991. Integrated rice tungromanagement in South Sulawesi. Indon. Crop Prot. 10:30–40.Savary S, Tiongco E, Fabellar N, Teng PS. 1993. A characterization of rice tungro epidemicsfrom historical survey data in the Philippines. Plant Dis. 77:376–382.Tiongco ER, Fabellar NG, Teng PS, Koganezawa H. 1992. Tungro viruses in volunteer riceplants. Int. Rice Res. Newsl. 17:20.Widiarta IN, Yulianto, Muhsin M. 1997. Distribution status of rice tungro disease in West Java.Indon. J. Plant Prot. 3:23–31.NotesAuthors’ address: Hasanuddin, I.N. Widiarta, and Yulianto. Research Institute for Rice, JI. RayaIX Sukamandi 41256, Subang, West Java, Indonesia.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management. 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.Improving IPM technology 137

Leafhopper control by insecticides is notthe solution to the tungro problemS. VillarealGreen leafhoppers (GLH) are one of the most abundant canopy arthropodsin irrigated rice systems throughout much of South and Southeast Asia.Populations are rarely large enough to cause direct feeding damage torice, but in some areas they are important pests as vectors of rice tungrodisease. In some countries in Asia, chemical control of GLH based onthreshold numbers of the insect is still recommended. In the majority ofrice-growing areas, however, tungro disease is absent or occurs infrequently.Consequently, routine insecticide applications against GLH cannot be justified.Moreover, spraying insecticides to control GLH does not always resultin effective tungro disease management. Insecticides may be harmfulto human health and to the environment and indiscriminate use of certaincompounds in rice has been shown to cause outbreaks of secondary pests.Alternative methods of managing tungro disease in endemic areas areavailable and these should be used for GLH control. Currently, virus-resistantvarieties are being developed and it is hoped that they will soon beavailable to rice farmers and that these varieties will provide them with afurther option for managing tungro.The main leafhopper vector of tungro disease, Nephotettix virescens (Distant), can befound in almost all irrigated and rainfed rice systems in tropical Asia. In a studyconducted at five irrigated sites in the Philippines in 1989, N. virescens was the mostabundant phytophagous insect in all locations (Heong et al 1989). In the absence ofsources of inoculum, however, this insect does not harm the rice crop through directfeeding. There have been reports of “hopperburn” caused by large populations ofGLH, especially from India, but these cases are extremely rare. In most rice-growingareas. inoculum sources are few or absent so that attempting to control leafhopperswould be wasting farmers’ time and money.GLH abundance and tungro disease incidence in endemic areasIn studies conducted in experimental plots on the IRRI farm in 1991-91, there was nocorrelation between numbers of N. virescens and tungro disease (Chancellor et al1996a). In one season. high tungro incidence uas recorded in the presence of lownumbers of leafhoppers. The availability of inoculum was considered to be the keydeterminant of the disease level. Once primary inoculum had been introduced intofield plots. rapid secondary plant-to-plant spread occurred. An analysis of factorsaffecting tungro incidence in endemic areas in Mindanao, Philippines, however, revealedthat increases in disease incidence were associated with rising vector numbersas well as with the number of viruliferous vectors (Savary et al 1993). Similar findingswere recorded in an endemic area in South Luzon, Philippines, during a survey

conducted in 1993-94 (Chancellor et al 1996b). In view of the impracticability ofreducing vector numbers to levels needed to significantly reduce tungro spread, however,the most appropriate tungro management strategy in such areas would be tominimize the amount of initial inoculum by planting resistant varieties or by reducingthe variation in planting dates.Efficacy of insecticide sprays against GLH in managing tungroMany field trials have been conducted on research stations to evaluate the efficacy ofvarious insecticides against GLH and their effectiveness in reducing tungro disease.Chancellor et a1 (1997) reviewed the results from some of these trials and concludedthat successful control of GLH and tungro disease can be achieved through insecticidesprays but that this may not be possible under certain conditions. A modelingstudy by Holt (1996) provides an explanation for this. In his study, the author examinedthe effect on final tungro incidence of applying a single insecticide at 20 d aftertransplanting, assuming an extremely high mortality rate of 95% to N. virescens. Themodel output indicated that, at very low immigration rates of GLH, a reduction indisease incidence of up to 60% could be achieved. As the immigration rate of infectivevectors increased, however, the reduction in tungro incidence became extremelysmall. Although this case refers to a situation with a single insecticide application, theassumed mortality rate of 95% is unrealistically high and this level of mortality couldprobably only be achieved with the use of two or more insecticides.Batay-an and Mancao (this volume) report on an insecticide trial carried out inNorth Cotabato, Philippines, in which good control of GLH and tungro disease wasachieved using synthetic pyrethroids. Where there are widespread sources of inoculumand active movement of GLH between fields, however, such an approach maynot be effective. Similar conditions were observed during a tungro epidemic in Negros,Philippines, in the 1998 wet season, where farmers resorted to weekly sprays ofcypermethrin in an attempt to control the disease. The strategy was not effective inreducing tungro incidence and the sprays caused the resurgence of populations of thebrown planthopper, Nilaparvata lugens (Stål), which led to “hopperburn” (Tiongco1998). Similarly, in an insecticide trial conducted in North Cotabato in 1997, weeklyspraying was not successful in preventing tungro from spreading in experimentalplots. Tungro disease incidence in surrounding rice fields was very high and presumablythere was continuous recolonization of the trial plots by viruliferous leafhopperssimilar to the process described by Schoenly et al (1996).Environmental and health concerns associated with insecticideapplications in riceThe cartoon illustrated in Figure 1 was used in training courses for rice farmers ontungro management conducted in India and the Philippines in 1996-98. The cartoonillustrates two of the unwelcome effects of spraying insecticides in rice. First, insec-140 S. Villareal

Fig. 1. Cartoon used for tungro management training courses in India and the Philippines showingthe negative effects of insecticide application for leafhopper control.ticide application methods in rice are still very primitive and, with the types of sprayersand nozzles available, the efficacy is doubtful. Most people doing the spraying donot wear protective clothing and they mix the chemical without wearing gloves. Theyspray in front of them so that they have to walk through the sprayed area. Unfortunately,category 1 and 2 insecticides are still widely available in many Asian countriesso the risk to human health through insecticide spraying is serious. Second, mostinsecticides that are used for insect control in rice are still nonselective and are harmfulto the many predators and parasites that are so important for regulating populationsof insect pests in rice.ConclusionsApplying insecticides to control leafhopper vectors of tungro disease cannot be justifiedin areas where inoculum sources are not present. Even in endemic areas, insecticideapplications are often not effective. In a survey conducted in 1994 in the Philippines.some farmers in tungro-endemic areas said that they continued to apply insecticidesto control tungro even though they knew that this approach did not work(Warburton et al 1997). They sprayed because they did not know what else they coulddo and they thought the insecticides might help to control some other insect pests.Leafhopper control by insecticides 141

Our challenge is to promote the adoption of more effective and environmentally safetungro management strategies in such areas.ReferencesChancellor TCB, Cook AG, Heong KL. 1996a. The within-field dynamics of rice tungro diseasein relation to the abundance of its major leafhopper vectors. Crop Prot. 15: 439–449.Chancellor TCB, Tiongco ER, Holt J, Villareal S, Teng PS, Fabellar N. Magbanua MGM.1996b. Risk factors for rice tungro disease in endemic areas. In: Rice tungro disease epidemiologyand vector ecology: the development of sustainable and cost-effective pestmanagement practices to reduce yield losses in intensive rice cropping systems. IRRIDiscuss. Pap. Ser. No. 19. Manila (Philippines): International Rice Research Institute.Chancellor TCB, Heong KL, Cook AG. 1997. The role of leafhopper control in the managementof rice tungro disease. In: Chancellor TCB, Thresh JM. editors. The epidemiologyand management of rice tungro disease. Chatham (UK): Natural Resources Institute,Heong KL, Aquino GB, Barrion AT. 1989 Population dynamics of plant- and leafhoppers andtheir natural enemies in rice ecosystems in the Philippines. Crop Prot. 11:371–379.Holt J. 1996. Spatial modeling of rice tungro disease epidemics. In: Rice tungro disease epidemiologyand vector ecology: the development of sustainable and cost-effective pest managementpractices to reduce yield losses in intensive rice cropping systems. IRRI Discuss.Pap. Ser. No. 19. Manila (Philippines): International Rice Research Institute.Savary S, Fabellar N, Tiongco ER, Teng PS. 1993. A characterization of rice tungro epidemicsin the Philippines from historical survey data. Plant Dis. 77:376–382.Schoenly KG, Cohen JE, Heong KL, Arida GS. Barrion AT, Litsinger JA. 1996. Quantifyingthe impact of insecticides on food web structure of rice-arthropod populations in a Philippinefarmer’s irrigated field: a case study. In: Polis GA, Winemiller KO, editors. Foodwebs: integration of patterns and dynamics. New York: Chapman and Hall. p 343–351.Tiongco ER. 1998. Unpublished trip report to Negros, Philippines. August 1998. Nueva Ecija(Philippines): Philippine Rice Research Institute.Warburton H, Palis FL, Villareal S. 1997. Farmers’ perceptions of rice tungro disease in thePhilippines. In: Heong KL, Escalada MM, editors. Pest management of rice farmers inAsia. Manila (Philippines): International Rice Research Institute. p 129–141.NotesAuthors’ address: S. Villareal, International Rice Research Institute, MCPO Box 3127, MakatiCity 1271, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9–11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.142 S. Villareal

The role of vector control in rice tungrodisease managementN. WidiartaTungro disease, which is spread by rice green leafhoppers, especiallyNephotettix virescens, is one of the most destructive diseases of rice. Thedisease is successfully suppressed by planting rice at the recommendedtime to avoid high disease pressure and by rotating varieties with resistanceto green leafhopper in synchronously planted areas, for example, inSouth Sulawesi. Simultaneous planting, however, is difficult to practice byfarmers for various reasons. Therefore, until recently, tungro disease expansionand outbreaks have mainly occurred in asynchronously plantedareas, primarily in Bali and Java. The population density of green leafhoppers in paddy fields in those areas is maintained at a low level by thedispersal activity of adults. Integrated pest management (IPM) strategieshave been developed based on characteristic population dynamics of thevector to reduce the proportion of viruliferous vectors. In this study. theuse of antifeedants against N. virescens to control tungro spread in synchronouslyand asynchronously planted fields was also examined. Theantifeedant and virus transmission inhibition activities of andrographolide,a major compound of Andrographis paniculata, and commercial insecticidessuch as imidacloprid, pymetrozin, MIPC, and nytenpyram were testedagainst female adults of N. virescens. Imidacloprid and nytenpyram showedbetter antifeedant activities than andrographolide and pymetrozin.Imidacloprid inhibited acquisition and inoculation of tungro viruses betterthan the others. The results imply that antifeedants have a potential toreduce virus transmission without directly disturbing the food chain. Inasynchronously planted areas, application of diacloden and MlPC successfullydecreased vector population density but not tungro spread. In synchronouslyplanted areas, application of diacloden, imidacloprid, and MlPCsignificantly reduced both vector population density and disease incidence.Rice tungro disease (RTD) has affected almost all provinces of Indonesia. Recently,the affected area has extended to the northern coastal lowland of West Java in Subang(Hasanuddin et al 1995, Widiarta et a1 1997a). Therefore, the important West Javarice bowl is now threatened.RTD has been successfully controlled in Sulawesi by transplanting synchronouslyover wide areas at the recommended planting time (Sama et al 1991). This enablesrice plants to escape peak periods of disease pressure. Synchronous planting was alsocombined with rotation of varieties particularly resistant to the tungro vector,Nephotettix virescens.Rice is susceptible to tungro infection during the early vegetative stage. Infectionat this stage causes severe yield loss. The older the rice plant is when infected, thelower the yield loss. The recommended planting time is based on the annual cycle ofN. virescens populations and RTD incidence. In Maros, South Sulawesi, April andOctober are the months of highest tungro risk because population densities of greenleafhopper (GLH) and RTD incidence are high then. Therefore, farmers are recommendednot to have young rice plants in the fields at this time to escape damage.

The escape strategy is combined with varietal rotation between seasons. Breedershave identified seven genes with resistance to GLH. Rice varieties are categorizedinto five groups based on the resistance gene of the parent.After the combined escape and rotation strategy was implemented, the tungroinfectedarea in South Sulawesi decreased remarkably. Before synchronous plantingwas established in 1984, the tungro-infected area exceeded 4,000 ha; now it is lessthan 100 ha.The escape strategy is difficult to implement in asynchronously planted areasbecause it is difficult for farmers to plant simultaneously over wide areas. Varietalrotation reduced disease incidence in asynchronous areas as long as GLH did notadapt to the varieties being used. The problem is, however, that resistant varietieswith different resistance genes also differ in eating quality. Farmers need varietieswith diverse resistance genes but that arc similar in quality. Recommended varietiesmay not be accepted because of their inferior taste. Almost all Indonesians preferIR64-type quality.Following the success of tungro management in South Sulawesi, the main problemhas switched from Sulawesi and Bali in 1980-85 to Java and Bali more recently.Unfortunately, Java and Bali contribute more than 60% of the total rice production inIndonesia. Consequently, tungro threatens rice production and self-sufficiency in thecountry. Other control strategies are therefore required, especially for asynchronouslyplanted areas.Population dynamics of N. virescensThe fluctuating RTD incidence in asynchronously planted areas is closely related tochanges in GLH density (Suzuki et al 1992). RTD incidence increases when GLHnumbers increase. To establish a control strategy for RTD in asynchronously plantedfields, the population dynamics of GLH was studied intensively in Bali using aFARMCOP insect suction trap (Cariño et al 1979, Aryawan et al 1993). All arthropodscaught were identified in the laboratory. It was established that the population densityof GLH increased only at the early stages of rice growth and decreased thereafter(Widiarta 1992). Some studies showed that population density did not increase at allafter immigrants invaded rice fields. The peak density of N. virescens was much lowerthan that of its sibling species, N. cincticeps, in temperate paddy fields. The populationdensity of N. ciscticeps was observed in Okayama, Southwestern Japan, using aprocedure similar to that used in Bali (Widiarta 1993). The population density of N.cincticeps increases from the time it infests rice fields until just before harvest and itspeak densities are much higher than those of N. virescens.A key factor analysis for the life tables of N. virescens and N. cincticeps wasconducted. Key component mortality can be identified with reference to the biggestslope of regression coefficient. Results showed that nymphal mortality including lossof adults after emergence is the key factor for both N. virescens and N. cincticeps.Egg parasitism is the second key factor for N. virescens but not for N. cincticeps.144 Widiarta

To establish relationships between key component morality and biotic and abioticfactors, regression analsysis was conducted using nymphal mortality and predatorsas well as nymphal mortality and rainfall. Nymphal mortality for virescnes wasnot influenced by spider numbers.Nymphal mortality for N. virescens was not related to physical factors such asrainfall. Therefore, adult loss after emergence is considered to be the key fator.RTD control strategyThe previously mentioned characteristic population dynamics of N. virescens suggestthat the population density of N. virenscens in asynchronously planted fields iskept low by the dispersal activity of adults. Newly emerged adults have short residentialperiods within fields, probably because they move to other fields of younger riceplants. Therefore, they lay fewer eggs in fields where they emerged. Egg parasites alsoplay an important role in reducing egg survival. Therefore reducing the populationdensity of N. virescens is not the most appropriate way to control tungro. The requirementis to reduce the proportion of inefective vestors.Virus sources can be reduced by using resistant varieties that restrict virus multiplicationand by practicing selective weeding of alternative virus hosts. Antifeedantscan be used to reduce the duration of virus inoculation and acquisition. This reviewpresents the role of antifeedants in controlling vector feeding and of insecticides incontrolling vector density in relation to RTD.Control of N. virescens feeding activityThe use of an antifeedant reduces virus acquisition and inoculation without killingthe insect, which helps maintain the balance with its natural enemies, especially eggparasites. Some substances have been tested in the laboratory (Widiarta et al 1997b).Feeding on artificial diets using the streaked parafilm method was employed. Theinsects tested were allowed to feed through parafilm. The feeding rate and number ofstylet marks on the parafilm were observed. Insect survival after treatment was alsorecorded to discriminate between antifeedant activity and insecticide activity.Table 1 shows that andrographolide, an active compound of the tropical plantAndrographis paniculata, and pymetrozin insecticide supressed feeding by femalesat the lowest concentration of 20 ppm, while imidacloprid and nytenpyram, aneonicotinoid insecticide, did so at concentrations of 0.01 ppm. Insect survival ratesat concentrations tested were generally high, but decreased markedly at the highestconcentration of 40 ppm for both the andrographolide and pymetrozine. The chemicalstested showed antifeedant activity against N. virescens.Andrographolide at 40 ppm and the insecticides pymetrozine, imidacloprid, andnytenpyram significantly reduced the number of stylet marks (Table 2). The minimumfeeding acquisition and inocultaion times for transmission of tungro by N.virescens were reported to be 5–30 min (Wathanakul and Weerapat 1969). The reductionin numberof feeding marks may be used as an indicator of the extent to whichvirus transmission is impeded.The role of vector control 145

Table 1. Mean consumption of female Nephotettix virescens after feeding24 h on 5% sugar mixed with various concentrations of test materialsthrough parafilm.Test material Concentration Survival rate Consumption(ppm) (%) (mg) aAndrographolide 10 93 9.7 ac20 100 6.8 ab40 70 9.3 aPymetrozin 10 90 9.4 ac20 80 6.0 ab40 53 4.6 blmidacloprid 0.01 93 4.6 bNytenpyram 0.01 90 5.9 abSugar (5%) 100 19.1 ca Means followed by a common letter are not significantly different at the 95% level,Fisher-PLSD test.Table 2. Mean number of stylet marks by Nephotettix virescenson parafilm after 24 h of access to various concentrations oftest materials.Test material Concentration (ppm) Stylet marks aAndrographolidePymetrozinlmidaclopridNytenpyramSugar (5%)a Means followed by a common letter are not significantly different at the95% level, Fisher-PLSD test.1020401020400.010.0161.9 af75.9 af37.4 b17.9 cd25.6 bc15.0 d9.5 e21.5 c93.6 fRice tungro virus transmission inhibition by antifeedantsThe effect of antifeedants on virus acquisition and inoculation was investigated(Widiarta et al unpublished). RTD-affected plants were treated with antifeedant beforeN. virescens was given access to the plants. Healthy rice seedlings were alsotreated with antifeedant before infective N. virescens was given access to them toassess feeding inhibition. The test tube inoculation method was used for both tests.Results showed that pymetrozin and andrographolide applied to tungro-affectedplants significantly decreased the number of vectors that became infective (Fig. 1).Application of imidacloprid at 0.01 to 0.02 ppm completely prevented acquisition.Pymetrozin and andrographolide applications on test plants at concentrations of20 ppm significantly reduced virus transmission by N. virescens (Fig. 2). Further-146 Widiarta

Fig. 1. Influence of antifeedants on acquisition of rice tungro viruses by green leafhoppers C = control;P20 = pymetrozin 20 ppm; P40 = pymetrozin 40 ppm; A20 = andrographolide 20 ppm; A40 =andrographolide 40 ppm; l1 = imidacloprid 0.01 ppm; l2 = imidacloprid 0.02 ppm. Bars followed by acommon letter are not significantly different at the 95% level Fisher-PLSD test (Widiarta et al inpress).Fig. 2. Influence of antifeedants on the inoculation of rice tungro disease by infective green leafhopper.C = control; P20 = pymetrozin 20 ppm; P40 = pymetrozin 40 ppm; A20 = andrographolide 20 ppm; A40= andrographolide 40 ppm; I1 = imidacloprid 0.01 ppm; l2 = imidacloprid 0.02 ppm. Bars followed by acommon letter are not significantly different at the 95% level by Fisher-PLSD test (Widiarta et al inpress).The role of vector control 147

more, increasing the concentration of pymetrozin to 40 ppm reduced virus transmission,but increasing that of andrographolide did not. Application of imidacloprid atconcentrations of 0.01 and 0.02 ppm significantly reduced virus transmission. It wasclear that imidacloprid greatly reduced virus transmission by N. virescens, whileandrographolide and pymetrozin did so only at lower concentrations.Control of N. virescens populationA field experiment was conducted in the 1997 dry-season crop in asynchronouslyplanted fields in Subak Padanggalak, Bali, and in the wet-season crop for synchronouslyplanted fields in Subak Samsam, Bali (Widiarta et al unpublished). Insecticideswere applied fortnightly, starting 1 wk after GLH infestations occurred. Theinsecticides tested were diacloden, imidacloprid, and MIPC. The experiment wasconducted using a randomized block design with four replicates of each treatment.The plot size of each replicate was 5 × 8 m. The susceptible rice variety Pelita wastransplanted at 25 × 25-cm spacing, 21 d after sowing. The population density ofgreen leafhoppers was observed weekly by counting 39 hills plot -1 . RTD incidencewas observed at 8 wk after transplanting and before harvest.In the asynchronous plantings, applications of diacloden and MIPC significantlyreduced the population density of GLH (Fig. 3). RTD incidence before harvest, however,was not reduced significantly (Fig. 4). In the synchronous planting, applicationsFig. 3. Population development of green leafhopper in variously treated plots in asynchronouslyplanted fields. MIPC = active ingredient of Mipcin 50 WP, Ga = diacloden at 12.5, 25, 50, and 100ppm.148 Widiarta

Fig. 4. Tungro incidence before harvest in variously treated plots in asynchronously planted fields.Varietal bars with the same letters are not significantly different at the 5% level by Duncan’s multiplerange test. Ga = diacloden at 12.5, 25, 50, and 100 ppm, MIPC = active ingredient of insecticideMipcin 50 WP, Cont = control.of diacloden, imidacloprid, and MIPC significantly reduced not only GLH density(Fig. 5) but also RTD incidence (Fig. 6). Thus, reduced vector density decreased RTDincidence, especially in synchronously planted crops.ConclusionsLow doses of insecticides such as imidacloprid, nytenpyram, and pymetrozin as wellas andrographolide showed antifeedant activity against N. virescens and also inhibitedtransmission of rice tungro viruses. Thus, control of feeding has the potential tocheck tungro without disturbing the food chain. Population control of GLH by insecticidesshowed the limitation of this approach in reducing RTD spread in asynchronousplantings.The role of vector control 149

Fig. 5. Population development of green leafhopper in variously treated plots in synchronously plantedfields. MlPC = active ingredient of Mipcin 50 WP, Ga = diacloden at 25, 50, and 100 ppm, Conf =imidacloprid (Confidor 25 WP).Fig. 6. Tungro incidence before harvest in variously treated plots in synchronously planted fields.Bars with the same letters are not significantly different at the 5% level by Duncan’s multiple rangetest. Ga = diacloden at 25, 50, and 100 ppm, Conf = imidacloprid, Mip = Mipcin, Cont = control.150 Widiarta

ReferencesAryawan IGN, Widiarta IN, Suzuki Y, Nakasuji F. 1993. Life table analysis of the green riceleafhopper, Nephotettix virescens (Distant) (Hemiptera: Cicadellidae), an efficient vectorof rice tungro disease in asynchronous rice fields in Indonesia. Res. Popul. Ecol. 35:31–43.Cariño FO, Kenmore PE, Dyck VA. 1979. The FARMCOP suction sampler for hoppers andpredators in flooded rice fields. Int. Rice Res. Newsl. 1:21–22.Hasanuddin A, Widiarta IN, Yulianto. 1995. Status of rice tungro disease in Central Java andWest Java. A paper presented at the Annual Meeting of the Indonesian Society of Phytopathology,25–27 September 1995, Mataram, Indonesia.Sama S, Hasanuddin A, Manwan I, Cabunagan RC, Hibino H. 1991. Integrated rice tungrodisease management in South Sulawesi. Indon. Crop Prot. 10:34–40.Suruki Y, Widrawan IKR, Gede IGN, Raga IN, Yasis, Soeroto. 1992. Field epidemiology andforecasting technology of rice tungro disease vectored by green leafhopper. Jpn. Agric.Res. Q. 26:98–104.Wathanakul L, Weerapat P. 1969. Virus disease of rice in Thailand. In: Proceedings of a Symposiumon the Virus Disease of the Rice Plant, 25–28 April 1967, Los Baños, Philippines.Baltimore: Johns Hopkins Press. p. 79–85.Widiarta IN. 1992. Comparative population dynamics of green leafhoppers in paddy fields ofthe tropics and temperate regions. Jpn. Agric. Res. Q. 26:115–123.Widiarta IN. 1993. Comparative population dynamics of green leafhopper. Nephotettix virescensand N. cincticeps. Shokubutsu-boeki (Plant Protection) 47:396–39 (in Japanese).Widiarta IN, Yulianto, Muhsin M. 1997a. Distribution status of rice tungro disease in WestJava. Indon. J. Plant Prot. 3:23–31.Widiarta IN, Usyati N, Kusdiaman D. 1997b. Antifeedant activity of andrographolide andthree synthetic insecticides against rice green leafhopper. Nephotettix virescens (Distant)( Hemiptera: Cicadellidae). Bull. Plant Pests Dis. 9: 14–19.NotesAuthor’s address: N. Widiarta, Research Institute for Rice, J1. Raya IX Sukamandi 41256.Subang, West Java, Indonesia.Citation: Chancellor TCB, Azzam O, Heong KL. editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.The role of vector control 151

GLH control for the managementof rice tungro diseaseT. Ganapathy, N. Subramanian, and M. SurendranGreen leafhopper (GLH) is one of the major sucking pests of rice and iscapable of transmitting rice tungro disease (RTD), a devastating virus diseaseof rice. GLH is considered to be more important as a virus vectorthan as a direct pest of rice. In a seedbed protection trial, application ofneem cake in the seedbed followed by spraying of 5% neem seed kernelextract (NSKE) at 30 d after transplanting reduced the incidence of tungrodisease by more than 50% and increased rice grain yield. GLH abundanceand the presence of a higher percentage of transmitters resulted in highRTD incidence. A significant positive linear relationship was observed betweenRTD incidence and both log transmitters and percent transmitterpopulation during an RTD epidemic in the northern districts of Tamil Nadu,India, during 1991-93.Rice tungro disease (RTD) is caused by two morphologically, chemically, and serologicallyunrelated viruses—rice tungro bacilliform virus (RTBV) and rice tungrospherical virus (RTSV). The rice green leafhopper (GLH), Nephotettix virescens (Distant),is the most efficient vector and can transmit both viruses either together orseparately (Hihino et al 1979). RTBV, however, is acquired by GLH only after prioracquisition of RTSV.GLH does not usually cause any significant feeding damage to the rice cropunless its numbers are extremely high. GLHs that carry tungro viruses, however, maycause extensive damage and crop loss (Anjaneyulu et al 1994).Where an overlapping rice cropping pattern occurs, GLH prefer young seedlingsand migrate from older crops to seedbeds which can then become infected(Mukhopadhyay et al 1984). Tungro spreads very fast during the early growth stageof the crop (Shukla and Anjaneyulu 1981). Hence, it is essential that rice seedlings beadequately protected from tungro infection. This can be achieved by applying insecticidesin seedbeds to kill immigrant vectors to prevent virus transmission. A seedbedprotection trial was therefore conducted to study the effect of nursery protection onthe incidence of GLH and RTD.The onset of the disease depends on the presence of a susceptihle host, a virussource. and the vector. Anjaneyulu (1975) observed that adult GLH plays an activerole in introducing primary inoculum to the field, whereas both nymphs and adultshelp in further secondary spread. The availability of virus inoculum, a high populationof GLH, and early growth stage of the crop are responsible for a disease outbreak(Chen and Othman 1991). High vector populations (Vidhyasekaran and Lewin 1986)and a large proportion of viruliferous vectors (Savary et al 1993) also play an importantrole in disease outbreaks. A direct correlation between vector population anddisease incidence has been observed at different locations (Lim 1972, Hibino 1986,Shukla and Anjaneyulu 1981). The presence of viruliferous GLH is one of the mostimportant factors that cause RTD incidence.

Materials and methodsSeedbed protectionTwo field trials were conducted to study the effect of seedbed protection with pesticideson the incidence of RTD and yield. A susceptible variety, IR64, was used in thetrials, which relied on naturally occurring sources of virus inoculum. Seedlings wereraised in a nursery and transplanted into 8 × 8-m plots at 24 d after sowing (DAS).There were five treatments. Each treatment was replicated three times in a randomizedcomplete block design: T 1 = carbofuran at 6.25 g m -2 at 20 DAS, T 2 = carbofuranin seedbed at 20 DAS and monocrotophos (0.2%) at 25 d after transplanting (DAT),T 3 = neem cake at 1.5.5 g m -2 at 20 DAS, T 4 = neem cake in seedbed and 5% neem seedkernel extract (NSKE) at 25 DAT, T 5 = control.Numbers of GLH adults were estimated at 14, 28, and 56 DAT using 10 sweepsof a 30-cm-diameter insect net in each plot. Nymphal numbers were counted on 10hills in each plot. Rice plants were scored visually for tungro incidence on the samedates. After harvest, grain yield was recorded.Vector indexingA rice tungro epidemic occurred during 1991-93 in the northern districts of TamilNadu. Using data collected at the Rice Research Station (RRS) in Tirur, the relationshipbetween tungro incidence, monthly numbers of GLH collected in a light trap,and the proportion of virus transmitters was examined. The percentage of transmittersin the GLH population was calculated by vector indexing. This was done asfollows:Individual GLHs were placed in a test tube containing 7-10-day-old TN1 seedlingsfor a 24-h inoculation feeding period. Seedlings were subsequently planted inpots and grown in insect-proof cages. One week after inoculation, the number ofseedlings showing typical symptoms of RTD was counted. The percentage of seedlingsinfected was considered to be equal to the percentage of viruliferous GLH. Thefollowing formula was derived:Log transmitters (Lt) = log 10 (Tg × Pvv)where Tg = total monthly light trap GLH collection and Pvv = proportion of viruliferousvectors (% vv/10) where % vv = (no. of transmitters/no. of GLH tested) × 100The relationship of RTD incidence with log transmitters and percent transmitterswas examined using regression analysis.Results and discussionSeedbed protectionProtection of the seedbed with insecticides and biopesticides reduced tungro incidencecompared with the control (Table 1). Adult GLH numbers at 14 and 28 DAT154 Ganapathy et al

Table 1. Green leafhopper (GLH) population, percent rice tungro disease (RTD), and grain yield inseedbed protection trial at Vishar during July to October 1997.GLH count and RTD IncidenceTreatments a 14 DAT 28 DAT 56 DATYieldGLH b RTD GLH RTD GLH RTD (t ha -1 )Adult Nymph (%) Adult Nymph (%) Adult Nymph (%)T1T2T3T40.71.70.71.30.20.30.30.30.00.00.00.01.00.73.72.71.01.31.40.318.120.420.023.51.30.01.72.00.80.61.00.325.524.424.523.32.62.72.13.0T54.7 0.6 0.0 7.0 2.3 22.3 3.3 1.0 40.8 1.6a T 1 = carbofuran at 6.25 g m -2 at 20 DAS, T 2 = carbofuran at 6.25 g m -2 at 20 DAS + monocrotophos (0.2%) at 25 DAT,T 3 = neem cake at 15.5 g m -2 at 20 DAS, T 4 = neem cake at 15.5 g m -2 at 20 DAS + 5% neem seed kernel extract at25 DAT, T 5 = control. b adult = mean no. 10 sweeps -1 , nymphs = mean no 10 hills 1 , DAT = days after transplanting.were lowest in plots where carbofuran and neem cake were applied in the seedbed.but nymphal numbers were similar in all treatments. The highest grain yield wasrecorded in T 4 , where neem cake and NSKE were applied.These results are consistent with previous findings. Protection of rice seedlingsgrown in soil incorporated with 150 and 250 kg neem cake ha -1 was effective againstrice tungro (Saxena 1987). Also, NSKE and neem cake powder mixed with carbofuranat 1.0 kg ai ha -1 reduced tungro incidence similarly as applying a higher rate ofcarbofuran alone (Abdul Kareem et al 1988).EpidemiologyAt RRS, Tirur, viruliferous GLH were first observed in August 1991 and initial diseasesymptoms appeared in September when 16% of rice plants were affected. Thedisease was prevalent up to December 1991, with a peak incidence of 35% duringOctober, which corresponded to the highest percentage of transmitters. Even thoughthe GLH population increased from January to March 1992, no RTD incidence occurredbecause of the absence of transmitters. Again, the GLH population increasedin May 1992 and, because of the presence of transmitters, symptoms reappeared duringJune 1992. Numbers of viruliferous GLH and numbers of GLH per hill peaked inAugust and November 1992. Because of the presence of a high percentage of transmitters.RTD incidence reached a maximum of 96% during August 1992. RTD incidenceand the percentage of transmitters started declining at the end of 1992. Nofurther disease was recorded in February 1993.An analysis of the relationship between log transmitter population and RTD incidenceusing linear and exponential models revealed that the linear regression modelprovided the best fit to the data (r = 0.831, error mean square = 400.4) (Table 2). Therelationship between the percentage of transmitters (percent viruliferous GLH) andRTD incidence was also studied. A highly significant positive relationship betweenGLH control 155

Table 2. Relationship of RTD incidence with log and percenttransmitters at Tirur.Transmitters Parameters EMS ar a bLogLinear model bExponential modelPercentLinear modelExponential model0.8310.7740.9780.865-73.13-52.712.87-39.7844.48102.212.9933.21400.4483.2264.5381.8a EMS = error mean square.b Linear model: y = a + bx, exponential model: y= ae bx .the percentage of transmitters and percent RTD incidence was observed in the linearmodel (r = 0.978) (Table 2). Suzuki et al (1992) demonstrated a relationship betweenthe percentage of infective GLH and the percentage of RTD-infected hills in ricefields at 5-7 wk after transplanting. They developed an infective vector index andfound it useful in predicting cumulative infection at early stages of crop growth inIndonesia.ReferencesAbdul Kareem A, Boncodin MEM, Saxena RC. 1988. Neem seed kernel or neem cake powderand carbofuran granule mixture for controlling green leafhopper (GLH) and rice tungrovirus (RTV). Int. Rice Res. Newsl. 13(3):35.Anjaneyulu A. 1975. Nephotettix virescens (Distant) nymphs and their role in the spread of ricetungro virus. Curr. Sci. 44:357–358.Anjaneyulu A, Satapathy MK, Shukla VD. 1994. Rice tungro New Delhi: Oxford and IBHPublishing Co. Pvt. Ltd. 227 p.Chen YM, Othman AB. 1991. Tungro in Malaysia. MAPPS Newsl. 15(1):5–6.Hibino H. 1986. Epidemiology of rice tungro. In: Proceedings of the Workshop on Epidemiologyof Plant Virus Diseases, 6–8 August 1986, Orlando, Florida.Hibino H, Saleh N, Roechan M. 1979. Transmission of two kinds of rice tungro associatedviruses by insect vector. Phytopathology 69:1266–1268.Lim GS. 1972. Studies on penyakit merah disease of rice. III. Factors contributing to an epidemicin North Krian, Malaysia. Malay. Agric. J. 48:278–294.Mukhopadhyay S, Roy J, Raychudhuri R. 1984. Management of rice tungro virus disease inWest Bengal, India. Rev. Trop. Plant Pathol. 1:181–195.Savary S, Fabellar N, Tiongco ER, Teng PS. 1993. A characterization of rice tungro epidemicsin the Philippines from historical survey data. Plant Dis. 77:376–382.Saxena RC. 1987. Neem seed derivatives for the management of Nephotettix virescens (Distant)and rice tungro virus. In: Proceedings of the Workshop on Rice Tungro Virus. Ministryof Agriculture. AARD-Maros Research Institute for Food Crops. Maros, Indonesia.p 75–85.156 Ganapathy et al

Shukla VD, Anjaneyulu A. 1981. Spread of tungro virus disease in different ages of rice crop.J. Plant Dis. Prot. 88:614–620.Suzuki Y, Astika IGN, Widrawan IKR, Gede IGN, Raga IN, Soeroto, 1992. Rice tungro diseasetransmitted by the green leafhopper: its epidemiology and forecasting technology. Jpn.Agric. Res. Q. 26:98–104.Vidhyasekaran P, Lewin HD. 1986. Forecasting tungro (RTV) epidemics in Tamil Nadu. Int.Rice Res. Newsl. 11(6):36.NotesAuthors’ address: T. Ganapathy, N. Subramanian, and M. Surendran, Department of Plant Pathology,Tamil Nadu Agricultural University, Coimbatore 641003, Tamil Nadu, India.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management. 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.GLH control 157

Management of rice tungro disease bychemical control of the green leafhoppervectorE.H. Batay-An and S.C. MancaoThe field efficacy of five synthetic pyrethroid insecticides was evaluated forthe control of green leafhoppers (GLH) and rice tungro disease (RTD) andfor their effect on natural enemy populations and yield of IR64. In each oftwo seasons, pyrethroid-treated plots significantly reduced GLH populationsand had significantly lower RTD incidence. Among the five treatments,plots treated with cypermethrin and ethofenprox had the lowest GLH populationand RTD incidence at all sampling dates. Yields were higher in thetreated plots than in the control. Populations of spiders, Cyrtorhinuslividipennis, Conocephalus longipennis, Agriocnemis pygmaea, andcoccinellids were significantly affected by insecticide applications 1 d aftertreatment at 5, 20, and 35 d after transplanting (DAT). Cypermethrin, however,did not affect C. lividipennis populations at 20 DAT. Likewise, none ofthe insecticides reduced A. pygmaea populations at 35 DAT in the 1989wet season or populations of any of the natural enemies at 35 DAT in the1990 dry season. Sprays of cypermethrin and ethofenprox did not significantlyaffect the population of spiders, C. lividipennis, and A. pygmaea at20 DAT in the 1990 dry season.Mindanao is one of the largest islands located at the southernmost part of the Philippines.Large areas of rice are grown on the island, which has a relatively even distributionof rainfall throughout the year and is free from typhoons. One of the major riceproduction constraints in Mindanao is the occurrence of insect pests and diseases.These are brought about by favorable climatic conditions coupled with asynchronousplanting, continuous cropping of rice, and planting of susceptible cultivars. Thus, theyear-round availability of rice and the warm humid climate are conducive to insectproliferation and survival.Among the pests in Mindanao, rice tungro disease, which is transmitted by thegreen leafhopper, Nephotettix virescens, is the most destructive disease of rice. Outbreaksof RTD in the Philippines occurred in 1957, 1963, 1969, 1971, 1975, and 1977(Bergonia 1978). In Central Luzon, IR36 and IR42, both moderately resistant to greenleafhopper (GLH), were highly infected with RTD in 1984. Likewise, in Mindanao,high RTD incidences were observed in South and North Cotabato in 1985 and 1986and again in 1993-98 (Truong et al, “Rice tungro disease in the Philippines,” thisvolume).At present, the best way to prevent RTD is by using vector-resistant varietiesbecause there is still no variety resistant to the tungro virus. Continuous planting ofthese vector-resistant cultivars, however, may cause them to succumb to RTD becauseof the development of virulent GLH populations. Therefore, one possibility forreducing RTD incidence in the field is to control the insect vector with insecticides.Rice yield increases have been attributed to insecticide use under various conditionsin nearly all rice-growing countries (Lim and Heong 1984). Greater use of insecti-

cides by farmers is expected because other control strategies such as using resistantvarieties and intensifying the role of biocontrol agents take time and are limited inscope. Under laboratory conditions, some commonly used insecticides such as thosein the organophosphate and carbamate groups applied as foliar sprays caused 90-100% GLH mortality 24 h after treatment. They cannot prevent infection with tungroviruses (IRRI 1984, 1985), however, because of their slow effect—GLH can transmitvirus particles before they die. Thus, rapid-acting insecticides are needed. Syntheticpyrethroids offer some advantages because they are fast-acting and effective at lowdosages; thus, residue levels are likely to be low on crops (Ozaki et al 1984). TheMidsayap branch of PhilRice conducted this study to evaluate the field efficacy offive synthetic pyrethroid insecticides against GLH and RTD incidence. The objectiveswere to determine the best insecticide for GLH control, to assess the impact ofdifferent insecticides on natural enemies, and to determine the impact on the yield ofIR64.Materials and methodsThe experiment was conducted at PhilRice-Midsayap, Bual Norte, Midsayap, Cotabato,during the 1989 wet season (WS) and 1990 dry season (DS). It was laid out in arandomized complete block design consisting of six treatments with a plot size of 5 ×5 m replicated four times.The treatments were T 1 = plots sprayed with cypermethrin, T 2 = plots sprayedwith monocrotophos + cypermethrin, T 3 = plots sprayed with deltamethrin, T 4 = plotssprayed with lambdacyhalothrin, T 5 = plots sprayed with ethofenprox, and T 6 = untreatedcontrol.Pregerminated seeds of susceptible IR64 were uniformly sown on the seedbedraised under the dapog method and immediately covered with nylon mesh after sowingfor protection against birds and early insect pest infestation. Ten-day-old dapogseedlings were transplanted at 3 to 5 seedlings hill -1 spaced at 20 × 20 cm. Starting 5d after transplanting (DAT), three insecticide applications by knapsack sprayer weremade at 15-d intervals following the manufacturer’s recommended dosage with aspray volume of 300 L ha -1 .Fertilizer was applied at a rate of 60-0-0 kg NPK ha -1 . One-half of the nitrogenwas applied at planting whereas the remaining half was applied 5 to 7 d before panicleinitiation. Preemergence herbicide (butachlor) was applied at 3 DAT for weed control.GLH and natural enemy populations were estimated 1 d before and 1 d afterevery spray application by 10 sweeps of an insect net per plot. RTD incidence wasrecorded at 60 DAT by counting the number of infected hills per plot. Rice yield wastaken from a 10-m 2 harvest area per plot, threshed, and dried at 14% moisture content.Data were analyzed statistically using analysis of variance, and Duncan’s multiplerange test was used for comparison among treatments.160 Batay-An and Mancao

Results and discussionGLH populations were significantly lower in all plots treated with cypermethrin,monocrotophos + cypermethrin, deltamethrin, lambdacyhalothrin, and ethofenproxthan in the untreated control plots at all sampling dates in the 1989 WS (Table 1).Similarly, RTD incidence was significantly lower in the treated plots than in theuntreated control plot (Table 1). There was no significant difference between plotssprayed with cypermethrin, ethofenprox, and lambdacyhalothrin. This result is consistentwith findings reported from India (Satapathy and Anjaneyulu 1984, Krishnaiahand Ghosh 1990, Anjaneyulu and Bhaktavatsalam 1986) and from the Philippines(Macatula et al 1987).The populations of spiders, Cyrtorhinus lividipennis, Conocephalus longipennis,Agriocnemis pygmaea, and coccinellids were considerably reduced by pyrethroid insecticideapplications at 5 and 20 DAT in the 1989 WS (Tables 2-6). Cypermethrinand lambdacyhalothrin applications, however, had no effect on C. longipennis numbersat 20 DAT (Table 4). Likewise, none of the synthetic pyrethroids tested significantlyaffected the population of A. pygmaea at 20 and 35 DAT (Table 5).Synthetic pyrethroid insecticides applied at 20 and 35 DAT significantly reducedGLH numbers 1 d after spray application compared with the untreated control in the1990 DS (Table 1). At 5 DAT, only cypermethrin and ethofenprox significantlyreduced GLH numbers.Table 1. Green leafhopper population and % rice tungro disease (RTD) incidence as affected by foliarsprays of synthetic pyrethroid insecticides on IR64, PhilRice-Midsayap, 1989 wet season and 1990dry season.GLH (no. 10 sweeps -1 ) b% RTDlnsecticide a Application at 5 DAT Application at 20 DAT Application at 35 DAT incidence1989 wet seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1 DBT 1 DPT 1 DBT 1 DPT 1 DBT 1 DPT at60 DAT12.0 a16.3 a15.0 a9.7 a8.0 a21.7 a4.3 a5.0 a4.7 a6.7 a4.3 a5.3 a1.3 a4.7 a4.7 a4.0 a3.0 a37.7 b1.7 b6.0 ab9.3 a4.7 ab1.7 b11.0 a11.0 a 3.7 a11.7 a 6.7 a15.3 a 10.3 a17.3 a 6.3 a13.7 a 3.7 a36.0 b 34.0 b3.7 c 1.0 c2.0 c 1.0 c4.3 bc 1.7 c5.0 bc 2.0 c5.3 bc 1.0 c11.0 a 11.0 a10.3 a14.0 ab17.0 bc20.7 c8.7 a40.7 d3.7 ab2.7 b1.7 a6.7 bc9.3 c5.6 abc2.3 ab32.0 d1.0 c1.3 c4.7 ab 2.3 bc3.7 ab 2.7 bc1.7 b 1.3 c9.3 a 7.3 a18.0 a39.3 b42.0 b29.3 ab22.0 a72.0 ca Insecticides were applied by knapsack sprayer 3 times at 15-d intervals. Spray volume was 300 L ha-1 . b Average of 3replications. In a column, means followed by the same letter are not significantly different at 5% level Duncan'smultiple range test (DMRT). DAT = days after transplanting. DBT = days before treatment, DPT = days posttreatment.Management of rice tungro disease 161

Table 2. Field evaluation of foliar sprays of synthetic pyrethroids against spider populations on IR64,PhilRice-Midsayap, 1989 wet season and 1990 dry season.Spiders (no. 10 sweeps -1 ) bInsecticide a Application at 5 DAT Application at 20 DAT Application at 35 DAT1 DBT 1 DPT 1 DBT 1 DPT 1 DBT 1 DPT1989 wet seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl12.0 a11.3 a5.0 a5.7 a11.7 a21.7 a4.7 ab5.7 ab2.0 b3.3 ab7.3 a7.7 a3.7 b1.7 ab0.0 a0.0 a9.3 c14.0 d2.7 b2.0 b2.0 b3.0 b3.0 b9.0 a17.7 b18.7 b15.7 bc11.0 c26.0 a25.7 a8.3 ab4.0 b5.0 b5.3 b7.3 ab13.0 a7.3 a8.3 a6.0 a1.7 a11.0 a30.3 b4.7 bc2.7 c1.7 c2.0 c6.7 ab11.3 a12.3 c9.3 bc3.7 a6.3 ab14.7 c20.3 d8.0 a4.3 a5.7 a4.0 a5.3 a10.3 a4.0 ab7.0 b1.7 a2.3 a7.0 b16.3 c2.0 a2.0 a2.0 a1.7 a3.3 a3.3 aa Insecticides were applied by knapsack sprayer 3 times at 15-d intervals. Spray volume was 300 L ha-1. b Average of 3replications. In a column, means followed by the same letter are not significantly different at 5% level Duncan'smultiple range test (DMRT), DAT = days after transplanting, DBT = days before treatment, DPT = days posttreatment.Table 3. Field evaluation of foliar sprays of synthetic pyrethroids against Cyrtorhinus lividipennispopulations on IR64, PhilRice-Midsayap, 1989 wet season and 1990 dry season.Cyrtorhinus lividipennis (no. 10 sweeps -1 ) bInsecticide a Application at 5 DAT Application at 20 DAT Application at 35 DAT1 DBT 1 DPT 1 DBT 1 DPT 1 DBT 1 DPT1989 wet seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos +cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl16.3 ab19.0 abc10.7 a12.0 a10.0 a9.0 a5.3 a7.3 a6.7 a6.0 a10.0 a9.0 a9.7 b7.3 ab2.0 a2.3 a5.3 b11.7 a2.3 bc0.7 bc1.7 bc0.3 c5.3 b11.7 a12.0 ab8.7 a11.0 ab7.0 a4.7 ab5.7 a4.0 ab2.3 ab1.0 b3.0 ab4.7 ab5.7 a11.3 b8.3 b3.7 a3.0 a3.0 bc5.7 ab2.0 bc0.0 c0.0 c1.0 c3.0 bc5.7 ab16.3 cd8.3 ab4.3 a7.0 a2.3 a2.0 a2.3 a1.7 a1.7 a0.3 a2.3 a2.0 a9.3 b4.3 ab1.7 a6.0 ab0.3 a1.7 a0.0 a0.0 a0.0 a0.3 a0.3 a1.7 aa Insecticides were applied by knapsack sprayer 3 times at 15-d intervals. Spray volume was 300 L ha- 1 . b Average of 3replications. In a column, means followed by the same letter are not significantly different at 5% level by Duncan'smultiple range test (DMRT). DAT = days after transplanting, DBT = days before treatment, DPT = days posttreatment.162 Batay-An and Mancao

Table 4. Field evaluation of foliar sprays of synthetic pyrethroids againstConocephalus longipennis populations on IR64, PhilRice-Midsayap, 1989 wet sea-son and 1990 dry season.Conocephalus longipennis (no. 10 sweeps - 1 ) bInsecticide aApplication at 20 DATApplication at 35 DAT1 DBT 1 DPT 1 DBT 1 DPT1989 wet seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl5.0 ab3.3 a4.7 a2.7 a3.3 a9.3 b0.3 b0.0 b1.3 a0.7 b0.0 a0.3 a5.0 ab1.3 a3.7 a2.7 a3.0 a11.0 b0.0 b0.0 a0.3 a0.0 b0.0 a1.0 a4.7 ab 1.0 a3.0 a 0.7 a7.0 abc 2.0 a6.3 abc 1.3 a7.3 bc 3.7 a10.0 c 10.3 b2.7 a 1.3 a0.7 a 0.7 a1.0 a 0.0 a0.7 a 0.0 a3.3 a 2.3 a1.7 a 1.3 aa lnsecticides were applied by knapsack sprayer 3 times at 15-d Intervals. Spray volume was 300 Lha -1 . b Average of 3 replications. In a column, means followed by the same letter are not significantlydifferent at 5% level Duncan's multiple range test (DMRT). DAT = days after transplanting, DBT =days before treatment, DPT = days post treatment.Table 5. Field evaluation of foliar sprays of synthetic pyrethroids againstAgriocnemis pygmaea populations on IR64, PhilRice-Midsayap, 1989 wet seasonand 1990 dry season.Agriocnemis pygmaea (no. 10 sweeps -1 ) bInsecticide a Application at 20 DAT Application at 35 DAT1 DBT 1 DPT 1 DBT 1 DPT1989 wet seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl4.0 a4.3 a4.7 a3.0 a4.0 a7.3 a8.0 a9.0 a9.7 a9.0 a9.0 a8.7 a2.3 a2.7 a2.3 a1.0 a3.3 a7.7 b5.7 bc4.3 c4.3 c5.1 c6.3 bc9.0 ab4.0 a 2.3 a3.0 a 2.3 a3.3 a 0.7 a2.0 a 1.7 a2.3 a 2.0 a5.0 a 5.3 a4.7 a 1.7 a7.0 a 0.7 a4.0 a 0.7 a5.3 a 0.3 a5.0 a 0.3 a4.0 a 1.7 aa Insecticides were applied by knapsack sprayer 3 times at 15-d intervals. Spray volume was 300 Lha -1 . b Average of 3 replications. In a column, means followed by the same letter are not significantlydifferent at 5% level Duncan’s multiple range test (DMRT). DAT = days after transplanting, DBT =days before treatment, DPT = days posttreatment,Management of rice tungro disease 163

Table 6. Field evaluation of foliar sprays of synthetic pyrethroids against coccinellidbeetle populations on IR64, PhilRice-Midsayap, 1989 wet season and 1990 dryseason.Coccinellid beetle (no. 10 sweeps -1 ) bInsecticide a Application at 20 DAT Application at 35 DAT1 DBT 1 DPT 1 DBT 1 DPT1989 wet seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl1990 dry seasonCypermethrinMonocrotophos + cypermethrinDeltamethrinLambdacyhalothrinEthofenproxControl3.0 a4.0 a2.0 a5.7 a3.3 a5.7 a0.3 b0.0 b0.0 b0.7 b2.7 a0.7 b0.7 a0.0 a0.0 a0.0 a2.0 a6.3 b0.0 b1.7 ab0.0 b0.0 b2.7 a2.7 a4.7 a6.7 a10.0 a9.7 a5.0 a12.3 a2.3 a1.7 a2.3 a0.7 a1.3 a4.0 a2.3 a0.0 a1.0 a1.0 a2.7 a8.7 b0.0 a0.3 a0.0 a0.0 a0.3 a0.7 aa Insecticides were applied by knapsack sprayer 3 times at 15d Intervals. Spray volume was 300 Lha -1 . b Average of 3 replications. In a column, means followed by the same letter are not significantlydifferent at 5% level Duncan's multiple range test (DMRT). DAT = days after transplanting, DBT =days before treatment, DPT = days posttreatment.Table 7. RTD incidence (%) and yield (t ha -1 ) of IR64 as affectedby foliar sprays of insecticide, PhilRice-Midsayap, 1990 dry season.Insecticide aRTD incidence (%) b Yield c (t ha -1 )60 DAT 1 DPTCypermethrinMonocrotophos + cyperrnethrinDeltamethrinLambdacyhalothrinEthofenproxControl12.1 b19.0 b23.0 b19.1 b16.8 b74.4 a3.6 a2.9 a2.7 a3.2 a3.3 a1.9 ba lnsecticides were applied by knapsack sprayer 3 times at 15-d Intervals.Spray volume was 300 L ha -1 . b Average of 3 replications. In a column, meansfollowed by the same letter are not significantly different at 5% level byDuncan's multiple range test (DMRT). c 2 × 5-m yield sample. DAT = daysafter transplanting, DPT = days posttreatment.Percent RTD infection obtained visually at 60 DAT was significantly lower intreated plots than in the untreated control in the 1990 DS. There were no significantdifferences among the five synthetic pyrethroids tested (Table 7).Yields obtained from plots sprayed with cypermethrin, ethofenprox,lambdacyhalothrin, monocrotophos + cypermethrin, and deltamethrin ranged from164 Batay-An and Mancao

3.6 to 2.7 t ha -1 . These yields were significantly higher than the 1.9 t ha -1 yield of theuntreated control. This result supports the findings of Macatula et al (1987).The application of five synthetic pyrethroid insecticides on IR64 by foliar sprayat 5 and 20 DAT considerably affected the populations of spiders. C. lividipennis, A.pygmaea, and coccinellids in the 1990 DS (Tables 2-6). Ethofenprox, however, didnot reduce the numbers of spiders and coccinellids, and A. pygmaea and C. lividipennisadults were not affected by most pyrethroid applications at 20 DAT.Moreover, none of the five pyrethroid insecticides tested significantly affectednatural enemy populations at 35 DAT (Tables 2-6).ConclusionsFive synthetic pyrethroid insecticides were evaluated for their effect on GLH numbers,RTD incidence, and natural enemy populations at PhilRice-Midsayap. The effectof insecticide application on the yield of IR64 was also assessed.The evaluation results are summarized as follows:1. Foliar sprays of cypermethrin, monocrotophos + cypermethrin, deltamethrin,lambdacyhalothrin, and ethnofenprox significantly reduced GLH populations andRTD incidence in both seasons.2. Among the five pyrethroids, cypermethrin and ethofenprox were the most effectivein lowering GLH numbers and percent RTD incidence.3. All plots treated with synthetic pyrethroids yielded significantly higher than theuntreated control in the 1990 dry season.4. In the 1989 wet season, foliar sprays of five pyrethroids significantly affected thepopulations of spiders, C. lividipennis, C. longipennis, A. pygmaea, andcoccinellids at 5 and 20 DAT, except for cypermethrin on C. longipennis at 20DAT. None of the pyrethroids affected the A. pygmaea population at 35 DAT.5. In the 1990 dry season, foliar sprays of cypermethrin and ethofenprox did notaffect the populations of spiders, C. lividipennis, and A. pygmaea at 20 DAT.Moreover, none of the pyrethroids significantly reduced natural enemy numbersat 35 DAT.The five synthetic pyrethroid insecticides tested were all effective against GLH.Synthetic pyrethroid insecticide application should be started within 5 DAT and repeatedat least two times at 15-d intervals to reduce GLH populations and RTD incidence.ReferencesAnjaneyulu A, Bhaktavatsalam G. 1986. Effect of synthetic pyrethroids on tungro incidenceand vector control. Int. Rice Res. Newsl. 11(6):15.Bergonia HT. 1978. Control measures to prevent tungro virus outbreak. Plant Prot. News 8:4–16.IRRI. 1984. Insecticide evaluation report for 1984. Entomology Department. Manila (Philippines):International Rice Research Institute.Management of rice tungro disease 165

IRRI. 1985. Insecticide evaluation report for 1985. Entomology Department. Manila (Philippines):International Rice Research Institute.Lim GS, Heong KL. 1984. The role of insecticides in rice integrated pest management. In:Judicious and efficient use of insecticides on rice. Manila (Philippines): International RiceResearch Institute. p 19–39.Krishnaiah NV, Ghosh A. 1990. Efficacy of ethofenprox in preventing rice tungro virus infection.Int. Rice Res. Newsl. 15(3):30.Macatula RF, Valencia SL, Mochida O. 1987. Evaluation of 12 Insecticides against green leafhopperfor preventing rice tungro virus disease. IRRI Res. Pap. Ser. 128. Manila (Philippines):International Rice Research Institute. 10 p.Ozaki K, Kassai T, Sasaki Y. 1984. Insecticide activity of pyrethroids against the green leafhopper,Nephotettix cincticeps Uhler. J. Pestic. Sci. 9:155–157.Satapathy MK, Anjaneyulu A. 1984. Use of cypermethrin, a synthetic pyrethroid in the controlof rice tungro virus disease and its vector. Trop. Pest Manage. 30(2):170–178.NotesAuthors’ address: E.H. Batay-an and S.C. Mancao, Philippine Rice Research Institute, Maligaya,Muñoz, 3119 Nueva Ecija, Philippines.Citation: Chancellor TCB, Azzam O, Heong KL, editors. 1999. Rice tungro disease management.Proceedings of the International Workshop on Tungro Disease Management, 9-11November 1998, IRRI, Los Baños, Philippines. Makati City (Philippines): InternationalRice Research Institute. 166 p.166 Batay-An and Mancao

Rice tungro disease management - IRRI books - International Rice ...

Create successful ePaper yourself

Delete template?

Save as template?