Project IP-4 - CGSpace Report... · 2.A. Characterization and Genetics of Resistance to Rice Blast, Sheath Blight and Grain Discoloration 51 2.B. Characterizing and Using Partial

Annual Report 2000Annual Report 2000

For Internal CirculationFor Internal Circulationand Discussion Onlyand Discussion Only

October 2000October 2000

Centro Internacional de Agricultura TropicalInternational Center for Tropical Agriculture

Project IP-4:Project IP-4:Improved Rice Germplasm for Latin AmericaImproved Rice Germplasm for Latin America

and the Caribbeanand the Caribbean

Annual Report 2000Annual Report 2000

For Internal CirculationFor Internal Circulationand Discussion Onlyand Discussion Only

October 2000October 2000

Centro Internacional de Agricultura TropicalInternational Center for Tropical Agriculture

Project IP-4:Project IP-4:Improved Rice Germplasm for Latin AmericaImproved Rice Germplasm for Latin America

and the Caribbeanand the Caribbean

Project IP-4. Improved Rice Germplasm for Latin America and theCaribbean. Annual Report 2000. Centro Internacional de AgriculturaTropical (CIAT). Cali, Colombia. October 2000.

TABLE OF CONTENTS

Page

Executive Summary (58 kb) 1Research Highlights in 2000 5Work Breakdown Structure 9Project Logical Framework 10

OUTPUT 1. ENHANCING GENE POOLS (156 kb)1.A. Rice Improvement Using Conventional Breeding and Gene Pools

and Populations with Recessive Male-Sterile Genes13

1.B. Developing Upland Rice for Small Landholders 281.C. Advance and Evaluation of Inter-Specific Gene Pools 381.D. Introgression of New Plant Type Genes into LAC’s Gene Pools 441.E. The Use of Anther Culture and in Vitro Culture for Enhancement of

Gene Pools48

OUTPUT 2. CHARACTERIZING RICE PEST AND THE GENETICSOF RESISTANCE (814 kb)

2.A. Characterization and Genetics of Resistance to Rice Blast, SheathBlight and Grain Discoloration

51

2.B. Characterizing and Using Partial Resistance for the Control of RiceBlast

75

2.C. Characterization of the Complex of Rice Hoja Blanca Virus and T.oryzicolus

95

2.D. Foreign Genes as Novel Sources of Resistance to Rice HojaBlanca Virus and Rhizoctonia solani

125

2.E. Characterization of Entorchamiento: A Complex of Polymixagraminis and Rice Stripe Necrotic Virus

136

OUTPUT 3. ENHANCING REGIONAL RICE RESEARCHCAPACITIES AND PRIORITIZING NEEDS WITH EMPHASISON THE SMALL FARMERS (223 kb)

3.A. FLAR and Economics of Rice Production Systems 1453.B. Rice Economics 185

ANNEX 1. PRINCIPAL AND SUPPORT STAFF (13 kb) 205

1

Annual Report 2000

IP-4 Improved Rice Germplasm for Latin America and the Caribbean

Executive Summary

The CIAT IP-4 project to improve rice germplasm for Latin America and the Caribbean(LAC) has been continuously changing throughout its three decades of existence to meetthe needs of the rice farmers. It places special emphasis on the small landholders, ourpartners and the external funding environment. The project is regional in focus. Ourcomparative advantage is the ability to address constraints and opportunities that areregional in scope yet distinctive to LAC. This executive summary reflects our strategy tosolve these constraints by highlighting some of the research that is being done and toindicate where there are additional needs and opportunities.

Output 1. Enhancing Gene Pools

The enhancement of gene pools is the heart of the rice project. This is a collaborativeeffort with partners throughout the world. Increasing the genetic diversity of commercialvarieties depends on having enhanced germplasm pools to use as parents in crosses. Incollaboration with IRRI, we are adapting the “new plant type” to conditions in Latin America.Conventional crosses were made and evaluated. To accelerate the process, we are alsousing anther culture to produce double haploids of the best lines and these are beingevaluated for plant type, grain quality, yield and adaptation to pests, diseases and nonbiological constraints, like cold and iron toxicity. During the last year, the anther culturelaboratory at CIAT produced more that 7,200 double haploids that are being used in manyof the breeding efforts at CIAT as well as by FLAR and national programs.

Wild species are another source to broaden the genetic base of rice germplasm.Unfortunately crosses with wild species also introduce many undesirable traits. Widecrosses are useful to map areas with potential desirable traits in the rice genome becausethe parents have a greater number of differences that can be detected using moleculartechniques. In collaboration with WARDA many interspecies crosses including O. sativaBg90-2/ O. rufipogon were made, and together with the CIAT project SB-2 and Cornell atleast 25 putative quantitative trait loci (QTLs) of important agronomic traits were identified.These efforts are creating novel germplasm pools that will be the basis for much of the newdiversity in the commercial varieties of the future.

Recurrent selection is the systematic selection of desirable individuals from a populationfollowed by recombination of the selected individuals to form new populations. The malesterile trait is used to enhance the cycle process of recurrent selection. This is one of themajor strategies of the rice project to produce enhanced gene pools. CIAT and CIRADhave a history of collaboration, and two CIRAD rice scientists are stationed at CIAT. Manyenhanced populations are under development and this process is so flexible that the samepopulations often become the basis for enhanced gene pools for many different localities.These enhanced gene pools are designed to overcome specific constraints. One

2

germplasm pool was developed for cold tolerance in upland rice using a source fromMadagascar. This gene pool was targeted for development of varieties for small uplandfarmers. Another of the germplasm pools is selecting materials the are adapted to thealtillanura of Colombia. Many different types of enhanced gene pools were sent this yearfor evaluation by NARs and NGOs.

Output 2. Characterizing Rice pests and the genetics of resistance

There are several pest and disease problems that are unique to Latin America. Theseinclude specific races of rice blast and Rhizoctonia, rice hoja blanca virus (RHBV),Tagosodes orizicolus and a disease recently introduced from Africa into Latin Americacrinkling disease or entorchamiento. During this last year there were increasing reports ofother pests and diseases such as Sarocladium, a fungus disease causing sheath rot. Thechallenge of the CIAT rice project will be to broaden it scope and to promote integratedpest and disease management solutions.

Developing varieties with durable resistance to rice blast has long been an elusive goal.CIAT and its partners using selection in hot spot breeding sites have increased significantlythe level of resistance to rice blast and this is reflected in some of the commercial varieties.Still there are reports of breakdown of the blast resistance and more progress is needed. Amolecular understanding of the host/pathogen interaction is critical to develop moresystematic breeding methods.

This year a new method to evaluate the genetic structure of the blast fungus wasestablished and is being transferred to our partners in the region. There is additionalevidence that the combination of the blast resistance genes Pi-1, Pi-2 and Pi-11 shouldconfer durable resistance. In collaboration with CIRAD, additional blast resistance genesare being identified. More than 50 potential donors of blast resistance genes are identifiedfor use in breeding programs. Applying recurrent selection to develop gene pools withpartial resistance that can be useful to develop lines with durable resistance is anotheractivity in which progress is reported. We are encouraging efforts to identify localpopulations, know the reaction to the resistance genes and use molecular methods inconjunction with hot spot breeding to finally develop a large number of commercial varietieswith durable resistance.

The complex of rice hoja blanca virus and the planthopper T. orizicolus have been a focusof CIAT research for many years. Nevertheless until recently most of the commercialvarieties were highly susceptible to the hoja blanca disease. With improved screeningmethodologies each year there are varieties being release with increasing levels ofresistance to both the virus and the insect. One new variety and several advanced lineshave more resistance to RHBV than the principal source of virus resistance. This isprobably due to increased resistance to the vector. Today Colombian and Venezuelan ricefarmers have more germplasm options, but there still is a need to promote varieties forother countries. The research this year changed the focus from developing additionalcontrol and screening techniques to the understanding of the mechanisms of resistance toRHBV and T. orizicolus. This is being coupled with efforts to develop molecular markers to

3

specific components of resistance and to a more systematic approach to developing highlyresistant varieties.

Crinkling disease or entorchamiento is a complex of the fungal vector Polymyxa graminisand rice stripe necrotic virus (RSNV). This problem has spread throughout Colombia duringthe last decade and now is in Panama. It is expected that this will soon be a regionalproblem. In collaboration with FEDEARROZ, both management practices and germplasmdevelopment are being investigated to control the disease. This year, the methods ofevaluating germplasm continued to be improved. A highly reproducible method ofscreening germplasm in the greenhouse has been developed and many commercial andadvanced lines were screened. The are moderately good sources of resistance in the ricegermplasm, and O. glaberrima appears to be an even better source. We are beginning todevelop enhanced germplasm pools with multiple source of resistance to entorchamiento.

CIAT has been developing transgenic rice with resistance to RHBV for several years. Thisyear permission to begin field experiments was granted by the Colombian government andthe transgenic rice with greenhouse resistance to RHBV are being tested in the field. Thisis also allows for the first time, the screening of large populations. This was the secondyear of collaborative research with Rutgers University to develop transgenic rice withresistance to multiple diseases using a novel gene (PAP). Transgenic plants wereproduced, they are expressing the protein and are in the process of being evaluated toresistance to sheath blight.

Output 3. Enhancing Regional Rice Research Capacities and Prioritizing Needs withEmphasis on the Small Farmers

CIAT is a founding member of FLAR, and we have a special relationship with them. Beingdriven by the specific requests of the FLAR members, they must respond quickly to theirneeds. Therefore this is an important avenue for the CIAT rice project to understand theneeds and perspective of the farmers. CIAT and FLAR have to continue to explore ways tocollaborate more effectively to help ensure the success Latin American rice farmers.

A regional study was made of the yield gap. This is the difference between the potentialyield of a variety and actual yields at the farm level. This shows that there is a potential toincrease rice production approximately 20% without much additional cost to the farmers. Ifpractices are adopted to reduce the yield gap, Latin American rice will become even morecompetitive in the global market. This year FLAR members released 12 varieties. There isa perception that the research and new varieties benefits the larger more affluent ricefarmers. The study concludes that within the same agro-ecosystem farmers regardless ofsize, land tenure, type of technology, and social variables such as age and experience arelikely to obtain similar yields. This suggests that rice technologies are neutral, relativelysimple and quickly adopted. This means that the small rice farmers are benefiting from thetechnologies generated by CIAT and our partners.

4

Conclusion

The IP-4 rice project has had a very successful year with significant research advances atall levels from basic research to applied research in collaboration with a wide range ofpartners. This is laying the foundation for future varieties with higher and more stableyields, better quality, durable resistance to biotic and abiotic stresses. Our past efforts arebearing fruit as the countries throughout the region release new varieties and are onceagain producing enough rice to fulfill the demands of the growing populations. With thelarge resource base of land with adequately water, Latin American countries have thepotential to become major rice exporters. For this to happen, the research that will allowincreasing yields and global competitiveness must continue.

CIAT has as its mission the alleviation of poverty and a special obligation to assure that theproducts of our research are reaching the small poor farmers. Rice is a crop where theimproved technologies have had a highly beneficial effect by directly increasing the incomeof all the rice farmers, both large and small. This is contributing to economic growth ofregions where small farmers also obtain off farm income associated with the expansion ofthis crop. The rice project is seeking to strengthen the alliances that will increaseopportunities to have more direct impact with the poorest rice farmers. We have already setout the basis to work with all our partners to make this a reality.

5

Project IP-4 Annual Report 2000

Research Highlights

Output 1. Enhancing gene pools

1.A. Rice improvement using conventional breeding and gene pools and populationswith recessive male-sterile genes.

• CIAT and CIRAD lines for upland conditions were released as varieties in Brazil asBONANÇA and CARISMA; in Bolivia a CIRAD line was released as JASAYE and inColombia a CIAT line “Línea 30” is to be released next year.

• In China FCRI/YAAS released the variety YUNLU 29, from a cross between a Chineseline and IRAT 216.

• Four cycles of upland population enhancement by mass recurrent selection for bothsexes for resistance to rice Hoja Blanca virus and blast, and major agronomic traits wasmade for the populations PCT-4, PCT-A, and PCT-5. These enhanced populations arebeing used for line development.

• Two cycles of recurrent selection based on S2 line evaluation were completed for thePopulation PCT-4.

• Creation of a formal group named “Grupo de Mejoramiento Genético Avanzado deArroz” (GRUMEGA) coordinated by CIAT/CIRAD and EMBRAPA.

• Site-specific composite populations are being developed in collaboration withorganizations in Venezuela, Chile, Uruguay, Brazil, and China.

• Fixed line development by Partners in Colombia (Fedearroz), Argentina (University ofLa Plata and Corrientes, Chile (INIA Quilamapu), Venezuela (Fundación Danac) andCuba (IIA).

• One cycle of population enhancement by recurrent selection in Chile of populationPQUI-1 for cold tolerance and grain yield, and in Venezuela of population PFD-1 forgrain yield and quality is completed.

I.B. Developing upland rice for small landholders

• Two special projects were obtained to continue research on the upland rice for hillsidesfor small landholders.

• A new collaboration with partners of Central America has been developed and thepartnership with China has been reinforced.

• The first recurrent population with narrow genetic basis of upland rice for hillsides hasbeen provided to 7 country partners.

I.C. Advance and evaluation of inter-specific gene pools

• Parallel AB-QTL studies suggest that some regions of the rice genome harbor genes forimprovement of cultivated rice.

• Data indicate that QTLs derived from O.rufipogon are expressed in different geneticbackgrounds and environments.

6

• Twenty five putative QTLs derived from O.rufipogon affecting several importantagronomic traits n phenotype in Bg90-2/O.rufipogon have shown consistent grain yieldthrough different generations. They were identified in Caiapo/O.rufipogon cross whilst22 in the Bg90-2/O.rufipogon.

• Breeding lines selected based on phenotype in Bg90-2/O.rufipogon have shownconsistent grain yield through different generations.

I.D. Introgression of new plant type genes into LAC’s gene pools

• New plant type traits have been introgressed in germplasm from Latin America. Bothconventional lines and double haploid lines with the traits fixed by anther culture havebeen selected and are showing promising results.

I.E. The use of anther culture and in vitro culture for enhancement of gene pools

• Total of 7,200 doubled haploids lines were generated from rice anther cultured for thevarious breeding efforts stationed at CIAT. Three hundred R2 lines and nine hundredR3 lines were distributed this year to national program in Latin America and FLAR.

OUTPUT 2. CHARACTERIZING RICE PESTS AND THE GENETICS OF RESISTANCE

2.A. Characterization and genetics of resistance to rice blast, sheath blight and graindiscoloration

• The combination of the blast resistance genes Pi-1, Pi-2, and Pi-11 was identified asthe most appropriate to confer durable resistance to all genetic lineages and virulencespectrum of the pathogen.

• A blast nursery with more than 50 potential donors of resistance genes was initiated forroutine testing of stability and distribution in the region.

• A rep-PCR technique for the characterization of the genetic structure of blastpopulations was established and is being transferred to blast researchers in the region.

• The wild rice Oryza rufipogum and the Asian rice cultivar Remadja were identified aspotential sources of resistance genes to sheath blight.

• Seven sources of high levels of resistance to grain discoloration including five uplandand two irrigated lines were identified.

• A backcross program for the introgression of blast resistance genes into rice cultivarswas initiated.

2.B. Characterizing and using partial resistance for the control of rice blast

• Seventeen F5 lines with cold resistant, very early, and with a parent with a high partialresistance to rice blast disease, were selected by the FLAR, because they wereimmune to rice blast in Santa Rosa hot spot trials.

7

• A complete cycle of recurrent selection for the partial and complete resistance to riceblast disease, and for other agronomic traits was evaluated and shows promisinggenetic progress for resistance to rice blast.

• The first five recurrent populations with a narrow genetic basis, for the different types ofrice cropping systems, have been sent to 10 country partners in Latin America, Africa,Asia and Europe.

• A new collaboration with partners of Central America has been developed and thepartnership with China has been reinforced.

2.C. Characterization of the complex of rice hoja blanca virus and T. orizicolus

• New varieties with high levels of resistance to RHBV and T. orizicolus were released inColombia and Venezuela during the last year. This is giving the farmers more optionsand should contribute to prevent outbreaks of RHBV and lower the use of pesticides.

• Two new varieties and some advanced lines have higher resistance to RHBV thanColombia 1, which is considered the prime source of resistance to the virus. Thisdemonstrates the progress that is being made through conventional breeding.

• Progress is being made to understand the mechanisms of resistance to RHBV and T.orizicolus. This should lead to more systematic development of germplasm pools.

• Highly resistant varieties have multiple types of resistance to T. orizicolus and RHBV.• It was determined that the variety Fedearroz 50 has both antixinosis and antibiosis to T.

orizicolus. Although biotypes often develop overcome the resistance, it is possible thatcomplex resistance of Fedearroz 50 will make it a durable variety.

• This year a study was made to determine if there are T. orizicolus biotypes. Usingbiological traits and molecular methods, it was concluded that there are no distinctbiotypes of T. orizicolus in Colombia.

2.D. Foreign genes as novel sources of resistance to rice hoja blanca virus andRhizoctonia solani

• A total of 421 selected transgenic lines representing various generations, and F2populations derived from crosses with different varieties will be planted in the field onNovember 2000. These lines will be evaluated for RHBV resistance and agronomictraits following International as well as the Colombian environmental biosafetyregulations at Palmira experimental station. The approval for field testing by theColombian Biosafety Committee was issued on September 2000.

• Rice was successfully transformed with the RHBV NS4 gene.• A total of 35 independent transgenic events carrying the PAPI deletion mutant gene,

and 50 independent transgenic events carrying the PAPII gene had been generated upto now. A first set of plant tissue was sent to Rutgers University this summer foranalysis and plants with PAP gene expression were identified based on Westernanalysis. PAP expressing plants will be evaluated for sheath blight resistance undergreenhouse conditions.

8

2.E. Characterization of entorchamiento: a complex of Polymyxa graminis and ricestripe necrotic virus

• A greenhouse method that gives consistently high levels of disease pressure wasdeveloped and is being used to screen rice germplasm for resistance toentorchamiento.

• Wild species especially Oryza glaberrima appear to be the best sources of resistanceand it may be possible to develop rice varieties that are immune to entorchamiento.

Output 3. Enhancing Regional Rice Research Capacities and Prioritizing Needs withEmphasis on the Small Farmers

3.A. FLAR and economics of rice production systems

• A total of 12 varieties were released this year by FLAR members.• A study based on panel data obtained from 180 rice producers over the 1991-98 period

in Colombia revealed that the new technologies have been readily adopted by all typesof farmers.

• Within highly homogeneous edapho-climatic conditions (at the municipality level),farmers are likely to obtain very similar yields. Yield is independent of the rice grower'sindividual skills, the type of technology and social variables such as age, experience infarming and in rice production and level of education; it is also independent of farm sizeor type of tenure.

3.B. Rice Economics

• The producers' classification as good or bad performer is variable, and it differs fromsemester to semester. The changes observed in performance most likely respond tovariables that cannot be controlled and that randomly affect the rice operation. There donot seem to be farmers with consistently “superior” packages, practices andperformance, suggesting that rice technologies are neutral, quite universal, relativelysimple, and easily and quickly adopted.

9

Project IP-4. Improved Rice Germplasm for Latin America and the Caribbean

Project Goal

To improve the nutritional and economic well-being of rice growers and low income consumers in Latin America and theCaribbean through sustainable increases in rice production and productivity

Project Purpose

To increase rice genetic diversity and enhance gene pools for higher, more stable yields with lower unit production costs thatpropiciate lower prices to consumers and reduce environmental hazards

Enhancing Gene Pools Characterizing Rice Pests andthe Genetics of Resistance

Enhancing Regional Rice ResearchCapacities and Prioritizing Needs

with Emphasison the Small Farmers

10

Project Log-Frame

Narrative Summary Measurable Indicators Means of Verification Important AssumptionsGoal

Germplasm of beans, cassava, tropicalforages, rice and their wild relativescollected, conserved and enhanced andmade accessible to NARS and otherpartners.

• A sufficient number of accessions (of beans,cassava and tropical forages) representinggenetic diversity are conserved and managedex-situ.

• Strategies and guidelines for in-situmanagement of biodiversity of beans, cassavaand tropical forages have been developed andtested with users.

• Accessible germplasm of beans, cassava,tropical forages and rice meet NARS standardsin terms of productivity, stability, agronomictraits and user needs.

• Techniques and relevant information for moreefficient and reliable germplasm improvementare accessible to users.

• CIAT’s germplasm bankinventories.

• Partners technical reports.

• Annual reports.

PurposeTo increase rice genetic diversity andenhance gene pools for higher, more stableyields with lower unit production costs thatpropiciate lower prices to consumers andreduce environmental hazards.

• Evaluations of yield potential (interspecific,wide, elite crosses and recurrent selection).

• Continued use of improved germplasm byNARS.

• Monitoring rice production practices andmarkets.

• IPM practices in place for stable productionand cleaner environment.

• Rice lines selected with desired gene traits.• Potential sources for high levels of biotic and

abiotic stress resistance.

• Databases.

• Project, CIAT and NARS annualreports.

• Publications.• Promotional Activities

(conferences, training,workshops, field days)

• Stability (internal and external)

• National policies favor adoption of newtechnology.

Outputs1. Enhancing Gene Pools.2. Characterizing Rice Pests and the

Genetics of Resistance.3. Enhancing Regional Rice Research

Capacities and Prioritizing Needs withEmphasis on the Small Farmers

• Pathogen/pest variation and source ofresistance identified.

• IPM strategies.

• Workshops.• Training courses.• Farmers’ surveys.

• Project progress report for2000.

• Publications.• Project progress and

workshop reports

• Continued support fromCIAT/CIRAD/FLAR.

• Continued adequate funding.• Recommendations adopted by NARS and

implemented by farmers.

11

Narrative Summary Measurable Indicators Means of Verification Important Assumptions

OUTPUT 1. ENHANCING GENE POOLS

Activities

A. Rice improvement using conventionalbreeding and gene pools/populationswith recesive male-sterile genes.

- Evaluation of savannas upland ricelines in Latin American countries.

B. Developing upland rice for smalllandholders

C. Advance and evaluation of inter-specific gene pools.

D. Introgression of new plant type genesinto LAC's gene pools.

E. The use of anther culture and in vitroculture for enhancement of gene pools.

• Rice populations developed and improved(tolerance soil acidity; resistance to blast,RHBV, T. orizicolus (13); good grain quality;early maturity.

• Number of field trials planted and linesselected.

• Populations distributed to NARS for linedevelopment.

• Populations developed (14); populations inprocess (12); populations yieldtested/molecular characterized (4). Partners(WARDA, CIRAD, EMBRAPA, CORNELL).

• Number of crosses made (433); tropicalirrigated (226), temperate (155), upland (52).Number of selected lines.

• Double haploids: interspecific crosses (386 ),acceleration breeding populations (815),somaclones (3758-Venezuela; 4440-Colombia)

• Project progress report for2000.

• Field visits and evaluations intesting sites.

• Breeding populationsdistributed to LAC.

• Breeding populations instorage and field.

• Best lines and QTL’Sidentified.

• Breeding populations instorage and field.

.• Double haploids in storage

• Publications.

• Continued support fromCIAT/CIRAD/FLAR.

• Adequate funding and timely release ofbudget.

• Favorable climate.

• Continued financial support for antherculture lab.

• Crosses, field support and operationalcosts provided by FLAR.

12

Narrative Summary Measurable Indicators Means of Verification Important AssumptionsOUTPUT 2. CHARACTERIZING RICEPESTS AND THE GENETICS OFRESISTANCE

Activities

A. Characterizing the interactions of hostplant resistance to rice blast, sheathblight and grain discoloration

B. Characterizing and using partial andcomplete resistance for the control ofrice blast.

C. Characterizing the interactions of hostplant rice hoja blanca virus and T.orizicolus complex

D. Foreign genes as novel sources ofresistance to rice hoja blanca virus andRhizoctonia solani

E. Characterizing the interactions of hostplant, Polymyxa graminis and rice stripenecrotic virus that causesentorchamiento.

• Virulence spectrum and genetic structure ofrice pathogens.

• Molecular markers associated and numberof resistance genes.

• Sources of complete, complementary andpartial resistance.

• Rice lines with diversified resistance toRHBV and T. orizicolus.

• Understanding components of resistance tothe RHBV complex.

• Crop management components developed.• Transgenic lines with RHBV-viral genes with

reduced symptoms produced and evaluated.• Transgenes introgressed into commercial

cultivars.• Using novel genes for multicomponent

resistance to rice pathogens.• Characterization of the RSNV and vector

complex.• Development of germplasm evaluation

methods.

• Collection of rice pathogens.• Database of resistance

sources• Crosses made among

resistance sources.• F7 lines with stable blast

resistance combining genesPi-1 and Pi-2.

• Rice genome map with blastresistance genes mapped.

• Rice progress report for 2000• Publications• Publication and diagnostic kit

available.• Resistant germplasm

selected under artificialconditions.

• Rice crosses and populations developedby breeders.

• Biotech. Unit identify molecular markersassociated with resistance.

• Continue collaboration with FLAR.• Continue adequate funding from

Colombia and Rockefeller.

• Continue support and adequate fundingfrom CIAT, CIRAD, and FLAR.

• Continued funding from Colombia,Rockefeller, Colciencias.

• Permission for field testing of transgenicplants is granted.

• Continued support and adequate funding.

OUTPUT 3. ENHANCING REGIONAL RICERESEARCH CAPACITIES ANDPRIORITIZING NEEDS WITH EMPHASISON THE SMALL FARMERS

Activities

A. FLAR and economics of rice productionsystems

- Analysis of national rice samples inColombia.

- Creation of a network of riceeconomics in Latin America (RECAL).

- FLAR breeding and crop managementactivities in LAC (training).

- Promotional and diffusion of activitiesand research impact.

B. Rice Economics

• Costs and coefficients of production.• National breeding plans written.• Number of scientists trained.• Published reports of courses.• FLAR publications.• Budget.

• Rice progress report for 2000. • Special funds continue.• Recommendations adopted by farmers.• Adequate funding and timely release of

budget.

13


1.A. Rice Improvement, Using Conventional Breeding and Gene Pools andPopulations with Recessive Male-Sterile Gene

1.A.1. Conventional Breeding for the Upland Savannas EcosystemM.Châtel, Y.Ospina, F.Rodriguez and V.H.Lozano

1.A.1.1. Introduction

As it has been stated in previous reports, we are gradually phasing out most ofthe activities involving the development of fixed lines for direct release to NARS.Since we have advanced segregating lines in the pipeline, we continue theirevaluation in Colombia and with our partners.Furthermore we also evaluate and select inter-specific progenies between Oryzasativa and O. glaberrima from WARDA and Oryza sativa lines from Madagascar.

1.A.1.2. Line Selection in Colombia

In Colombia, we continue line evaluation and selection to develop fixed materialto be tested by CORPOICA in the “Altillanura” conditions.CIAT/CIRAD conventional breeding linesIn 1999, 156 lines (26 families of 6 lines) were evaluated and 16 (10.2%) wereselected at , “La Libertad Experimental Station” (LES), Villavicencio – Meta. Ineach selected line 6 individual plants were harvested. Seeds were stored in thecold chamber, and a sample of each line sent to CIRAD.WARDA inter-specific lines and O. glaberrima accessionsIn 1999 one (1) WARDA line and three (3) O.Glaberrima accessions wereselected. Results evaluation were sent to Dr. Cesar Martinez.

1.A.1.3. Line Release in Brazil

Participation of CIAT/CIRAD material in different trials continues to be veryimportant. The main characteristics praised by Brazilian from CIRAD/CIATmaterial are earliness, plant and grain type. In 1999, 2 lines from CIAT, CT11614-1-4-1-M (CNA8172) and CT 11251-7-2-M-M (CNA8305), were releasedby EMBRAPA, as BONANÇA and CARISMA, respectively.

1.A.1.4. Line Release in Bolivia

In 1999 CIRAD line IRAT 170 was released as JASAYE by CIAT Santa Cruz.It is well adapted to rice cropping system used by small farmers.

1.A.1.5. Line to be Released in Colombia

“LÍNEA 30” or CIRAD 409 will be released next year by CORPOICA. It hasalready been tested in different cropping systems in rotation with others crops

14

and in inter-crop with perennial species.

1.A.1.6. Use of CIAT end CIRAD Lines in China

CIRAD- CA has close links with FCRI/YAAS in the Province of Yunnan, and hadshipped many upland lines in recent years. After screening, they were used fordirect release or as parents. In 1999, FCRI/YAAS released variety YUNLU 29,coming from a crossing between a Chinese line and IRAT 216. Lines fromCIAT/CIRAD Project are under evaluation in different trials and sites. Progeniesfrom CIAT/CIRAD and Chinese parents are also evaluated.

1.A.2. Population Breeding for Upland Savannas EcosystemM.Châtel, Y.Ospina, F.Rodriguez and V.H.Lozano


The upland rice population breeding project, using recurrent selection, aims atadapting, developing and selecting tropical japonica gene pools and populations.Major characteristics we bred for savanna conditions are:

• Tolerance of soil acidity• Resistance to diseases; rice blast (Pyricularia grisea Sacc.)• Resistance to pests, mainly rice plant-hopper (Tagosodes orizicolus)• Good grain quality (translucent, long-slender grain)• Early maturity (total cycle about 115 days)

1.A.2.2. Population Breeding

Activities reported here were conducted during cropping season 1999 (1999 A),from April to September at LES.

1.A.2.2.1. Line Development from Recurrent Populations

During enhancement of gene pools and populations through recurrent selection,fertile plants are selected.This selection is the starting point for the development of promising fixed lines forvariety release and/or potential parents for our regional partners (Argentina,Brazil, Bolivia, Colombia, Venezuela and the Caribbean through CRID Net).

1.A.2.2.1.1. Generation S1

Generation S1 comes from stored seeds of fertile S0 plants selected at PalmiraExperimental Station (PES) during off-season (1999 B) from October to March, inpopulation PCT-4 used for Yolima Ospina’ thesis.• PCT-4\0\0\0 (original population)• PCT-4\0\0\2 (original population with 2 recombination)

15

• PCT-4\SA\1\1 (one cycle of selection for Acid Soil and 1 recombination)• PCT-4\SA\2\1 (one cycle of selection for Acid Soil and 2 recombination)• PCT-4\SA\3\1 (one cycle of selection for Acid Soil and 3 recombination)• PCT-4\SA\4\1 (one cycle of selection for Acid Soil and 4 recombination)• PCT-4\SA\1\1,SA\1 (two cycles of recurrent selection for Acid Soil)

During 2000 A, a total of 229 S1 lines were observed and selected at LES.Seventy-one (71) S1 lines were selected by breeders of six countries thatattended Upland Rice Workshop held in LES, during August 7-11. (Bolivia: 14;Brazil: 10; Colombia: 14; Cuba: 14; Venezuela: 5; Argentina: 14).


The S2 generation comes from S0 fertile plants selected from two populationsduring the cropping season 1999.• Population PCT-11\0\0\2 (Second cycles of recombination \2\)Cropping season 1999 A. Nineteen (19) individual S0 fertile plants were selected.During 1999 B, S1 generation was grown at PES. During 2000 A, S2 generationwas observed and selected at LES. Results are not yet available.• Populations PCT-4\0\0\0; PCT-4\SA\1\1; PCT-4\SA\2\1 and PCT-4\SA\3\1(Original population, one recombination cycle \1\ after one selection for acid soil\SA\; two recombination cycles \2\ after one selection for acid soil \SA\) and threerecombination cycles \3\ after one selection for acid soil \SA\, respectively).Cropping season 1999 A. Two hundred thirty seven (237) individual fertile plantswere selected.During 1999 B, S1 generation was grown at PES. During 2000 A, S2 generationwas observed and selected at LES. Thirty-seven (37) S2 lines were selected by3 breeders that attended the Upland Rice Workshop held in LES, during August7-11. (Brazil: 8; Colombia: 15; Venezuela: 14).


Generation S4 comes from fertile S0 plants selected at LES during the croppingseason 1998 A. Generations S1 and S3 were advanced during 1998 B at PES.Generation S2 was selected in LES during the cropping season 1999.• Populations PCT-5\PHB\1\0,PHB\1,PHB\1PCT-A\PHB\1\0,PHB\1,PHB\; and PCT-4\PHB\1\1,PHB\1,PHB\1 (Third cycle ofrecurrent selection for leaf blast P and Hoja Blanca HB).Cropping season 1999 AOne hundred and seven (107) S2 lines were evaluated and 3 selected (3%) intwo populations (no selection was made in PCT-5\PHB\1\0,PHB\1,PHB\1).During 1999 B, S3 generation was grown at PES. All lines were discarded.• Population PCT-4\SA\2\1(Second recombination cycle \2\ after one selection for acid soil \SA\)Cropping season 1999 ASeventy-three (73) S2 lines were evaluated and 5 were selected (6.8%) at LES.

16

In each selected line a different number of individual plants was harvested.During 1999 B, 23 S3 lines (3 families of 6 lines, 1 family of 2 lines and 1 family of3 lines) were grown at PES. All lines were discarded• Population PCT-11\0\0\1(First cycle of recombination \1 of the original population)Cropping season 1999 AEighty-four (84) S2 lines were evaluated and 4 selected (4.8%) at LES. In eachselected line a different number of individual plants was harvested. During 1999B, 22 lines S3 (3 families of 6 lines, 1 family of 4 lines) were grown off-season atPES, and 7 were selected. During 2000 A, S4 generation was observed andselected at LES. Six (6) S4 lines were selected by 3 breeders that attended theUpland Rice Workshop held in LES, during August 7-11. (Brazil: 1; Colombia: 3;Venezuela: 2)


Generation S6 comes from S0 fertile plants selected during 1997 A at LES.Generations S1, S3 and S5 were advanced off-season at PES during 1997 B,1998 B and 1999 B respectively. Generations S2 and S4 were observed andselected at LES during the cropping seasons 1998 A and 1999 A respectively.• Populations PCT-5\PHB\1\0,PHB\1; PCT-A\PHB\1\0,PHB\1 andPCT-4\PHB\1\1,PHB\1 (Second recurrent selection cycle for leaf blast P andHoja Blanca HB)Cropping season 1999 AThirty-five (35) S4 lines were evaluated at LES and 7 selected (20%) inpopulation PCT-4\PHB\1\1,PHB\1 (lines from two others populations werediscarded). In each selected line 6 fertile plants were harvested. During 1999 B,42 S5 lines (7 families of 6 lines) were grown off-season at PES. Thirty-eight lineswere selected.During 2000 A, S6 generation was observed and selected at LES.• Population PCT-4\SA\1\1(First cycle of recombination \1\ after one selection for acid soil \SA\)Cropping season 1999 AThree hundred ninety (390) lines S4 were evaluated and 44 (11%) selected atLES. In each line 6 fertile plants were harvested.During 1999 B, 264 S5 lines (44 families of 6 lines) were grown off-season atPES. 251 lines were selected.During 2000 A, S6 generation was observed and selected at LES.From a total of 289 S6 lines from two populations (PCT-4\PHB\1\1,PHB\1 andPCT-4\SA\1\1), 387 S6 lines were selected by breeders that attended the UplandRice Workshop held in LES, during August 7-11. (Bolivia: 61; Brazil: 52;Colombia: 133; Cuba: 47; Venezuela: 33; Argentina: 61). The number of selectedlines is higher than the total number of lines evaluated. This comes from themultiple selection of some lines by different breeders.

17


Generation S7 comes from S0 fertile plants selected during 1996 A at LES.Generations S1, S3 and S5 were advanced at PES during 1996 B, 1997 B and1998 B, respectively. Generations S2, S4 and S6 were observed and selected atLES during the cropping seasons 1997 A, 1998 A and 1999 A, respectively.• Populations PCT-5\PHB\1\0;PCT-A\PHB\1\0 and PCT-4\PHB\1\1 (First cycle of recurrent selection for leaf blast P and Hoja Blanca HB)Cropping season 1999 A. One hundred and fifty (150) S6 lines were evaluatedand 13 (8.7%) selected, (in population PCT-4\PHB\1\1 only), at LES. In eachselected line, 6 individual plants were harvested. S6 lines were not advanced atPES during 1999 B, but were evaluated for grain quality and tolerance toTagosodes orizicolus. 78 lines were selected. During 2000 A, generation S7 wasobserved and selected at LES.Fifty-four (54) S7 lines were selected by breeders of the six countries thatattended the Upland Rice Workshop held in LES, during August 7-11. (Bolivia: 4;Brazil: 20; Colombia: 15; Cuba: 3; Venezuela: 8; Argentina: 4).


Generation S9 comes from fertile S0 plants selected at LES during the croppingseason 1995 A. Generations S1, S3, S5 and S7 were advanced at PES during1995 B, 1996 B, 1997 B and 1998 B respectively.Generations S2, S4, S6 and S8 were observed and selected at LES during thecropping seasons 1996 A, 1997 A, 1998 A and 1999 A respectively.• Populations PCT-5\0\0\0, PCT-A\0\0\0, and PCT-4\0\0\1(Basic populations with no selection)Cropping season 1999 AThree hundred one (301) S8 lines harvested in PES were evaluated and 44selected (14.6%) from the two populations PCT-A\0\0\0, and PCT-4\0\0\1 at LES.In each selected line, 6 individual plants were harvested. S8 lines were notadvanced at PES during 1999 B, but were evaluated for grain quality andtolerance to Tagosodes orizicolus. A total of 307 plants were selected.During 2000 A, S9 lines were observed and selected at LES. Two hundredseventy two (272) S9 lines were selected by breeders of the six countries thatattended the Upland Rice Workshop held in LES, during August 7-11. (Bolivia:41; Brazil: 66; Colombia: 56; Cuba: 38; Venezuela: 30; Argentina: 41).• Population PCT-4\0\0\1>S2Cropping season 1999 ASeventy-two (72) selected S8 lines (12 families of 6 lines) were evaluated, and 6were selected (8%) at LES. In each selected line, 6 individual plants wereharvested. During 1999 B, 36 lines (6 families of 6 lines) were grown at PES,and all were discarded.• Populations PCT-5\0\0\0>S3, PCT-A\0\0\0>S3 andPCT-4\0\0\1>S3 (Basic populations. Plant selection in S3 Lines at PES, 1996 B)

18

Cropping season 1999 A. Twelve (12) lines were evaluated and 1 was selected(8.3%) at LES. In each selected line, 6 individual plants were harvested. During1999 B, 36 lines (6 families of 6 lines) were grown at PES, and all werediscarded.

1.A.2.2.2. Yield Trials

1.A.2.2.2.1. Grain Yield and Quality Trial Using Advanced Generations

Advanced generations are promising fixed lines that passed through all selectionprocess. Selection of best yielding lines showing excellent grain quality wasmade at LES and PES during 1999 A and B semesters.During cropping season 2000, a trial was set-up at LES with the best 23 lines.Results are not yet available, but visually, some tested lines seem to beat thebest check: “Línea 30”.

1.A.2.2.2.2. INGER-LAC Acid Soil Nursery

In 1999, regional INGER-LAC acid soil nursery called “Vivero Internacional deObservación para América Latina” (VIOAL suelos ácidos) was prepared anddistributed to our partners.

1.A.2.2.3. Upland Line Registration

CIAT does not register lines: when a specific line does well in a given country,the national institution of that country may decide to name and release it forcommercial cultivation.CIRAD has a mechanism by which breeders can register specific materials. Aline is named CIRAD, and is also given its local synonym, if it is a result ofcollaborative work.During 1998 we applied for the registration of 1 line, registered as CIRAD 409 or“linea 30”. In 1999, this line seeds were increased at PES.

1.A.2.2.4. Line Dispatch to NARS

During 1999 we dispatched seed samples (fixed lines from the S8 generation) toour main collaborators in LAC and abroad for evaluation and selection. Samplesof 385 lines were sent to:Argentina Universidad de TucumánBrazil EMBRAPA - Arroz e FeijãoBolivia CIAT- Bolivia, Santa Cruz de la SierraVenezuela INIA (former FONAIAP)The Caribbean CRID NetCuba IIAChina FCRI/YAAS – Province of Yunnan

CATAS - Province of Hainan

19

Results of evaluations and local selection will be presented next year.

1.A.2.2.5. Population Enhancement

CIAT Rice Project emphasizes pre-breeding activities, and concentrates on theenhancement of populations to be distributed to NARS. The strategy is todevelop and enhance gene pools and populations for well-targeted traits. Theyare, therefore, used as reservoir of promising lines and/or potential parents to bedeveloped by national breeding programs.In the first 3 years of the project, we concentrated on introducing, characterizingand selecting germplasm from Brazil (former joint project of EMBRAPA - Arroz eFeijão). From 1995 onwards, we focused on enhancing selected populations anddeveloping new ones.

1.A.2.2.5.1. Recurrent Selection Based on S2 Line Evaluation

• Population PCT-4\SA\2\1Cropping season 1999 ARecombination.Population PCT-4\SA\2\1 with one cycle of recurrence and recombined twice wasgrown at LES to obtain the third recombination identified as PCT-4\SA\3\1Multi-site evaluation of S2 lines.A set of 155 S2 lines from the first cycle of recurrence (PCT-4\SA\1\1) was sent toBrazil (EMBRAPA - Arroz e Feijão), Bolivia (CIAT Santa Cruz), and Venezuela(UNELLEZ) for evaluation and selection for line development. In Bolivia 23 lineswere selected and their S3 progenies evaluated during the cropping season1999/2000. Thirteen lines were selected.• Population PCT-4\SA\3\1After the first selection cycle for acid soil (SA), population PCT-4 was recombined3 times (\3\). The resulting population was grown at LES during 2000 A, and S0fertile plants selected. Generation S1 will be grown at PES during 2000 B andthe S2 generation evaluated at LES during the cropping season 2001 A.• Population PCT-4\SA\1\1Population PCT-4\SA\1\1 with one cycle of recurrence (SA\1) was submitted to asecond cycle of recurrent selection. The resulting enhanced population (PCT-4\SA\1\1,SA\1) was grown during the year 2000 at LES. A third cycle ofrecurrent selection started through the selection of S0 fertile plants. GenerationS1 will be grown at PES and the S2 evaluated during 2001 A at LES.Evaluation of the S2 generation from the populations PCT-4\SA\3\1 andPCT-4\SA\1\1,SA\1This evaluation will be made during the cropping season 2001 A at LES. Thegenetic progress obtained from one cycle of selection followed by 3recombination will be compared two cycles of selection, each one of them onlyrecombined once.

20

1.A.2.2.5.2. Mass Recurrent Selection for Both Sexes for “Hoja Blanca”,Blast, and Major Agronomic Traits• Populations PCT-4\0\0\1, PCT-A\0\0\0, and PCT-5\0\0\0In 1999, after completed 4 cycles of recurrence, the enhanced populations werestored in the CIAT cold chamber at PES. They would be dispatched to LACNARS (Argentina, Brazil, Bolivia, Colombia and CRID Net) as reservoir ofgenotypes for line development.

Results of the enhancement• Total resistance to leaf blastFrom the first cycle of selection a drastic reduction in the number of infectedplants occurs. The use of mass recurrent selection on both sexes was efficientfor enhancing 3 populations. At the same time we selected for total leaf blastresistance, and for good agronomic characters. The enhanced populations willbe used as reservoir for the development of fixed lines.• Resistance to rice “Hoja Blanca” virusS2 evaluation of enhanced populations showed that 97.2% have resistant andintermediate reaction to Hoja Blanca. Enhanced populations can be consideredas good reservoirs for the development of resistant fixed lines by LAC NARS.

1.A.2.2.5.3. Recombination of the Population PCT-11

A new population was developed during 1997 and 1998.Cropping season 1999The second cycle of recombination was completed at LES. Fertile plants ofsecond recombination (population PCT-11\0\0\2) were selected, (see generationS1 and S2 mentioned above).Cropping season 2000Population PCT-11\0\0\2 was recombined one more time at LES and identifiedas PCT-11\0\0\3

1.A.2.2.6. Registering new Upland Populations

In 1999 one (1) new germplasm was registered on request of our partner inChina.• PYN-1: Japonica population for upland-hillsides ecosystem (FCRI/YAAS,

Province of Yunnan - China).

1.A.2.2.7. Special Study: Genetic Progress, Population PCT-4

Principal objectiveMaster Degree Thesis of Yolima Ospina at the “Universidad Nacional dePalmira”.

21

Specific objectiveEvaluation of genetic progress for acid soil tolerance and different agronomiccharacteristics like flowering time, plant height and grain yield, throughrecombination cycles. Genetic progress after one selection cycle and effect ofdifferent recombination cycles.

MethodEvaluation of S1 lines in two contrasted plots for soil acidity.Statistical design was Augmented Federer blocks composed by S1 from fourpopulation and 6 checks (3 susceptible - CICA 8, CICA 9 y Oryzica Llanos 5 -and 3 tolerant - Oryzica Sabana 6, Oryzica Sabana 10 y CIRAD 409 - to soilacidity).Data were collected and are currently processed.

1.A.2.2.8. Relationships with CORPOICA

CORPOICA Regional 8 in Villavicencio is interested in increasing research onupland savanna rice. We had two meetings with Director, Dr. Jaime Triana. Onemeeting was in Villavicencio during visit of Dr. Pierre Fabre from CIRAD-CA. Dr.Hernando Delgado is newly in charge of the rice savanna project. He attendedthe Breeding Workshop organized in August 2000 where he selected 236 lines.

1.A.3. Population Breeding for Lowland RiceM.Châtel, Y.Ospina


Population Breeding for Lowland irrigated rice is developed in closecollaboration with FLAR partners in Latin America and CIRAD partners in Europeand Asia.

The breeding population project started by introducing to Colombia, differentgene pools and populations previously developed in Brazil by EMBRAPA - Arroze Feijão and CIRAD, and by CIRAD, in French Guyana.Germplasm was characterized at PES, and the best-adapted ones were used todevelop new populations. This resulted in three populations that were registeredin recurrent selection catalog as PCT-6, PCT-7, and PCT-8. This work wasconducted at CIAT in close collaboration with Drs. C. Martínez and E. P.Guimarães.A gene pool was also built up using the gene of male sterility WC 232*5-EARLY(male-sterile gene of TOX 1011-4-1). Gene pool was registered as GPCT-9. Asecond gene pool, developed by CIRAD for temperate climates, was registeredas GPIRAT-10. From late 1996 this basic germplasm was dispatched to ourregional partners and outside Latin America. It is the starting point to develop

22

breeding population in different countries.In 1999 the II International Workshop on Rice Recurrent Selection, held inGoiânia-Brazil, was the occasion for our partners to present updated informationon the use of population breeding. Written documents were made available to allparticipants. A book edited by Dr. Elcio Guimarães was published jointly byCIRAD, CIAT, EMBRAPA and Fundación DANAC.At the Workshop’ final plenary session, the formalization of the informal networkthat we have developed through the past years with our partners, was discussed.It was decided to create a formal group named “Grupo de MejoramientoGenético Avanzado de Arroz” (GRUMEGA) with two coordinators: Dr. ElcioGuimarães (EMBRAPA - Arroz e Feijão) and Marc Châtel (CIAT/CIRAD). Aproposal to FAO for funding raising was prepared and circulated through ourpartners. FAO will not fund the network but is acting as a facilitator for fundraising. Until now the “informal network” has worked, but with external funds weshould be able to develop more networking activities.

1.A.3.2. Development of Site-Specific Composite Populations

1.A.3.2.1. Venezuela

Two populations, PCT-6 and PCT-7, were selected as the best introducedmaterial to be used as sources of male-sterile background to develop two newlocal populations, identified as PFD-1 and PFD-2.Cropping Season 1999At CIAT Palmira the built-up of population PFD-2 was completed and the basicpopulation shipped to DANAC-Venezuela.

1.A.3.2.2. Argentina

Argentina is developing 3 local populations: PARG-1, PARG-2 and PARG-3Cropping season 1999. Population PARG-3 was developed at CIAT byintroduction 50% of variability from 6 new lines into population PCT-8. The newlocal population will be ready in late 2000.

1.A.3.2.3. Chile

Chilean population PQUI-1 was selected in two sites with climatic differences(Chillán and Colchagua). Two populations, identified as PQUI-1\Ch\0\1 andPQUI-1\Co\1\0 were sent to CIAT for completing the third cycle of recombination.Cropping season 1999.At CIAT Palmira, we used remnant seeds of the second recombination toperform the third cycle during the first semester of 1999.

23

1.A.3.2.4. Uruguay

Three (3) populations are currently being developed at CIAT: PURG-1, PURG-2and PURG-3.Cropping Seasons 1999 and 2000.Build-up of the populations for Uruguay is under way at CIAT Palmira.

1.A.3.2.5. Brazil

CIRAD-CA has developed two populations, 00EP and 00NP, from thepopulations IRAT MANA, and PCT-6, respectively. IRGA is developing a localpopulation by the introduction of 7 lines into the Chilean population PQUI-1.

1.A.3.2.6. China

The Food Crops Research Institute (FCRI), of Yunnan Academy of AgriculturalSciences (YAAS), have developed a local germplasm for irrigated high altituderice ecosystem using male sterility from CIAT/CIRAD japonica population PCT-5.Local Chinese germplasm was identified as GPYN-2. It is a Gene Pool becauseof its less than 50% participation in original population PCT-5, as source of malesterility.

1.A.3.3. Maintaining Composite Populations

Because we manage the catalogue for rice germplasm for recurrent selection, wealso have the responsibility to ensure presence of sufficient seed in thegermplasm bank. Because of sufficient availability of seed, no multiplication wasdone this year.

1.A.3.4. Distributing Germplasm

In 1999, we sent subtropical PCT-8 and temperate PQUI-1 populations to newpartners in Europe (Spain and France).

1.A.3.5. Registering New Populations

In 1999 four (4) new germplasms were registered on request of our collaboratorsfrom Chile, China and Venezuela. PQUI-1 Japonica population for irrigatedtemperate ecosystem (INIA-Quilamapu, Chile). GPYN-2 Japonica gene pool fortemperate ecosystem (FCRI/YAAS, Province of Yunnan - China). PFD-1 andPFD-2 Indica populations for tropical lowland ecosystem (DANAC – Venezuela).

24

1.A.3.6. Line Development Through Anther CultureM.Châtel, Y.Ospina, Z.Lentini and A.Mora

1.A.3.6.1. From Chilean Populations

To hasten the development of fixed lines for Chile, Chilean populations wereprocessed by anther culture at CIAT.Cropping season 1999Populations PQUI-1\Ch\2\0 and PQUI-1\Co\2\0 were grown at CIAT Palmira andplants processed at the CIAT’ anther culture laboratory. 325 R1 lines wereproduced in 1999. The R2 generation were advanced in Colombia during the firstsemester of year 2000, and then dispatched to Chile for evaluation and selectionduring the 2000 – 2001 Chilean cropping season.These lines R1 were also shipped to French Rice Center (CFR) in Arles-France,for evaluation during the cropping season 2000. Information received from Dr.Guy Clement from CIRAD in charge of the evaluation, stated that these linesshowed excellent early vigor.

1.A.3.6.2. From Crosses from Romania

Two crosses (OLTENITSA / RUBINO, and CRISTAL / L 203) for cold tolerance,grain quality, and yield potential were processed by the CIAT anther culturelaboratory.Cropping season 1999A total of 61 R1 DH lines was produced. Seeds R2 were shipped to Romaniaand also to France and Chile. Results of evaluation are not yet available.

1.A.3.6.3. From Crosses from Spain

On request of collaborative project Spain and CIRAD-CA, four crosses weremade at CIAT Palmira and then processed by anther culture. Generation F1 wasgrown during late 1999. In early year 2000, plants were processed by antherculture and seeds R1 sent to CIRAD.

1.A.3.7. Fixed Line Development by Partners

Both from basic and enhanced populations, fertile plants were selected and theirprogenies are evaluated. Good variability was found and we expect a promisingfuture. In Colombia (FEDEARROZ), Argentina (University of La Plata andCorrientes, Chile (INIA Quilamapu), Venezuela (DANAC) and Cuba (IIA), there isgreat progress in developing new advanced lines.

25

1.A.3.8. Population Enhancement

Colombia

Populations PCT-6, 7, 8 and GPCT-9 are enhanced for resistance to the HojaBlanca Virus. Enhanced population PCT-6HB, best adapted to the tropics, wasremitted to Dr. M. Valès as basic germplasm to select durable blast resistance.One cycle of selection for total and partial resistance was completed, and thesecond one initiated.

Chile

Population PQUI-1 was selected in two different climatic conditions (Northernand Southern sites of the rice-growing region). Two sub-populations werederived (PQUI-1\CO and PQUI-CH). Enhancement of population PQUI-1\CH forcold tolerance was started, and one recurrent selection cycle was completedthrough laboratory and field screening.

1.A.4. Training Activities, Conference Organization and WorkshopM.Châtel, Y.Ospina and E.Guimarães - EMBRAPA

1.A.4.1. Training

National Course

We are organizing jointly with IIA (Cuba) and EMBRAPA (Brazil), a NationalCourse on Rice Population Breeding using Recurrent Selection to be held inCUBA in July 2001.

Thesis

Edwin Blandón Arias. Student of the “Facultad de Ingeniería Agronómica,Universidad del Tolima”, Colombia.“Caracterización y Adaptación en las condiciones del Norte del Tolima depoblaciones de Arroz (Oryza sativa L.) de amplia y estrecha base genéticadesarrolladas con un gen de androesterilidad”.Yolima Ospina. Assistant of the CIAT/CIRAD projectMaster Degree Thesis at the “Universidad Nacional de Palmira”, Colombia.“Evaluation of genetic progress for acid soil tolerance and different agronomiccharacteristics”.

1.A.4.2. Conference Organization

Brazil – September 1999The second International Conference on Rice Recurrent Selection Breeding washeld in Goiânia, 21 – 24 September 1999, and organized by EMBRAPA, CIRAD,CIAT, and Fundación DANAC, with the support of FAO and FLAR. All our

26

partners from different countries attended the conference and presented updatedinformation on the use of population breeding. CIAT, EMBRAPA and FundaciónDANAC published all the communications in a book.

1.A.4.3. Workshop

In early 2000 we planed with EMBRAPA, the venue of an International UplandRice Workshop. The idea was have our main partners to come to Colombia andselect upland rice lines that could be useful for their breeding project.The Workshop took place in Villavicencio – Meta, during August 7-11, 2000.Participants from Argentina (Universidad National de Tucumán), Bolivia (CIAT –Santa Cruz), Brazil (EMBRAPA – Arroz e Feijão), Colombia (CORPOICA,CENICAFE and Universidad del Tolima), Cuba (IIA) and Venezuela (INIA –former FONAIAP) attended the Workshop.Each participant presented a brief report about the progress made by his/herrespective upland rice breeding project. At “La Libertad experimental Station”,line selection was made by each participant (selections are presented in Activity2.). Selected material was harvested and shipped to the respective countries.

1.A.4.4. Publications

• Book ChaptersIn “Avances en el Mejoramiento Poblacional del Arroz” (published in October2000 by CIRAD, CIAT, EMBRAPA and Fundación DANAC).Chapter 6: Mejoramiento poblacional de arroz irrigado con énfasis en el virus dela Hoja Blanca. Borrero. J; Châtel. M and M.Triana.Chapter 8: Seleccion recurrente en Arroz: situación actual y perspectivas enCuba. Polanco. R. P; Châtel. M and E.Guimaraes.Chapter 10: Mejoramiento poblacional en Uruguay: caracterizacion y desarrollode germoplasma. Perez de Vida. F. B; Châtel. M and J.Borrero.Chapter 11: Desarrollo de poblaciones de arroz en Argentina. Marassi. J. E;Marassi. M. A; Châtel. M and J.Borrero.Chapter 12: Desarrollo de poblaciones japonicas para Chile. Hernaiz. S;Alvarado. R; Châtel. M and J.Borrero.

• Reports

- Achievements of the Rice Collaborative Project between CIRAD-CA and CIATCIAT- CIO Strategic Alliance Meeting. Montpellier-France, 22-24 June, 1999.- Collaborative Project between CIRAD CIAT and FLAR. Population Breedingusing gene pools and populations with recessive male-sterile gene, andconventional breeding. Annual Report CIAT/CIRAD/FLAR - October 1999.

27

1.A.4.5. Presentations

France- CIAT/CIO Strategic alliance meeting. The CIAT/CIRAD rice collaborative projectMontpellier, June 1999- Diversité des partenariats et excellence scientifique. Montpellier, September1999Bolivia- Selección recurrente en Arroz. Proyecto CIAT/CIRAD. Santa Cruz de la Sierra;February 2000Argentina- Arroz de secano en Latinoamerica. Universidad de Tucumán-Argentina; March2000

1.A.4.6. Concept Note

• Proposal for an Inter-Center Project between CIAT/CIRAD andWARDA/CORAF (Upland Rice Recurrent Selection for West Africa)

In 1999 a Concept note was prepared and circulated to CIAT and CIRAD. BothInstitutions agreed to the proposal and contacts were made with WARDA (witchassumed the mandate of CORA). In mid-year 2000, Dr. Aart Schoonhoven senta message to DG of WARDA proposing this project.

1.A.4.7. Visits to CIAT

During year 2000, the CIAT/CIRAD project received the following visitors:Dr. Pierre Fabre, CIRAD-CA, Head of the Food Crops Program (CALIM).Dr. Elcio Guimaraes, Brazil (EMBRAPA Arroz e Feijão).Dr. Eduardo Graterol, Venezuela (Fundación DANAC).Dr. Carlos Gamboa, Venezuela (Fundación DANAC).Dr. Marta Nicosia, National University of Tucumán - ArgentinaDr. Javier Osorio, National University of Tolima – ColombiaDr. Argemiro Moreno, Colombia, CENICAFE.

28


1.B. Rice Genetic Resources Improvement for Latin America and theCaribbean - Subproject - CIAT-CIRAD Collaborative Project

1.B.1. Genetic Sources Reinforcement• Confronting Food Insecurity in the Hillsides

1.B.1.1. Introduction

During CIAT-CIO strategic alliance meeting in June 1999, it was decided to thathillsides upland rice would not be a core funded activity. Therefore, the mainactivity for the year 2000 was the search of new funds and associates toreactivate this hillsides rice program.

1.B.2. Search for FinancingM.Vales

With support of CIRAD-CA, Montpellier, France, two projects were obtained:• Project of Foundation Aventis (Aventis Foundation, 2000): "Upland RiceDevelopment of the Madagascar altiplano and the Colombian Andes". This threeyear project was written together with Dr. Jean-Luc Dzido, CIRAD, who works inthe National Institute of Rural Development of Madagascar (FOFIFA). Basicallyit is a transfer of rice lines and crop management method toward small uplandrice producers. Associate countries are Colombia, Madagascar and China(Province of Yunnan).• Ministry of International Affairs (MAE) of France Project:"To assure production of upland rice in Latin America, China and Madagascar".It is a one-year rice breeding project. Associate countries are Colombia, CostaRica, Madagascar and China (Province of Yunnan).A project has been submitted to the Colombian Institute for the Development ofScience and Technology "Francisco José de Caldas", COLCIENCIAS.

To strength our activities to reach resource poor farmers, a new project wassubmitted to the National Program of Agricultural Technology Transfer,PRONATTA, of Colombia. This three year project focuses on and is entitled:"Rice Crop Reactivation in Guapi and Timbiqui in the Pacific Caucana Coast ".

1.B.3. Search for New PartnersM.Vales

• Development of collaboration with University of Costa Rica.A trip to Costa Rica was made from April 23 to 25, 2000.The main objective was to strengthen our new collaboration with the Rice ProjectDevelopment (DESARROZ) of Costa Rica University (UCR). DESARROZ’ roleis to encourage and coordinate rice collaboration in UCR (from basic to most

29

applied research), and with other production chain actors (Chamber of seedproducers, Chamber of rice producers, etc.). For Costa Rica, upland rice is apotential user and the UCR is a strong associate that can support neighboringcountries.

• Collaboration Reinforcement with CIAT’s Project PE3 in Central America.Participation in Los SOLes workshop of CIAT’ Hillsides Program, April 25 –27,2000 in San Dionisio, Nicaragua, was the opportunity to reinforcecollaboration with CIAT’s PE3 Project. During this workshop, it was planned withDr. Miguel Ayarza, a trip to Honduras from May 3 to 4, 2000, to get acquaintedwith the rice crop potential conditions and to specify the technical content of ourcollaboration in upland rice blast resistance.

• Reinforcement of collaboration with Chinese Agronomic Sciences Academyof Yunnan Province (YAAS).

YAAS is an associate partner in two new projects (See Activity 2.1.). Due to lackof supplementary resources for irrigating water in the north of Yunnan, Chineseauthorities want to develop for this zone upland rice varieties with coldresistance. From August 24 to September 4, 2000, a trip to the north of theYunnan Province was made to define transfer objectives and selection for uplandrice for the highlands with respect to the producer and consumer preferences inthis zone. The type of rice they want is very different to the one preferred in LatinAmerica. The main crop problem is Blast.

Deputy Director of YAAS, Dr. Tao Dayun, made two proposals:- A new collaboration was proposed: a program for upland rice hybrids with

cold and Blast resistance. This collaboration will permit to benefit China'sleadership on rice hybrids; CIRAD's leadership in upland rice for highlandsand the expertise of the collaborative project CIAT-CIRAD in durable blastresistance.

- To facilitate the transfer selection methods, in particular for the hybridprogram, Mr. Peng Xu, Dr. Tao Dayun’s assistant, was nominated for trainingin Colombia on: plant-parasite reciprocal recurrent selection, retro-recurrentselection, field selection of partial resistance, recurrent selection for partialand complete resistance, and other agronomic traits.

1.B.4. Pro-Vitamin A Genes Introduction from Irrigated Japonica Varietiesinto Japonica Upland Varieties for the Hillsides.

M.Vales, Z.Lentini

Professor Ingo Potrykus is making progress on the IPR issues to deliver his linesfor pro-vitamin A genes. It is expected that these transgenic lines will be availablein 2001.

30

1.B.5. Crosses to Obtain F2M.Vales, J.Carabalí


To begin population line creation, crosses were made to obtain F2 populations.

1.B.5.2. Material and Methods

Most of progenitors are F5 lines (table 1.1).

Nineteen crosses (table 1.2) were made using James Taillebois’ crossing method(Taillebois and Castro, 1986; Surapong, 1991).

1.B.5.3. Results and Perspectives

F1 seeds were obtained. These seeds will be planted during semester 2000 B toobtain F2 populations which will be evaluated in Popayan’s sub station (1700 m)during semester 2001 A.

1.B.6. Formation of a New Population with Narrow Genetic Base M.Vales, J.Carabali


To obtain varieties for the medium term, populations with narrow genetic baseare being made for high altitude (Vales et al., 1998).

1.B.6.2. Material and Methods

To avoid a genetic drift in this high altitude population, progenitors with coldresistance are used (table 1.3). Therefore most of the progenitors are lines F5selected for their high fertility in Popayan (1700 m). Also lines that have as aprogenitor with Blast resistant source are used (table 1.4).

Two series of crosses (table 1.5) were programmed to introduce the androsterilitygene and 7 cytoplasms in existing population, and also to make geneticrecombination. Fourteen crosses were made using James Taillebois’ method(Taillebois and Castro, 1986; Sarkarung 1991).

1.B.6.3. Results and Perspectives

F1 seeds were obtained. To improve genetic recombination, back crosses will bemade next semester, 2000 B.

31

1.B.7. Delivery of First Recurrent Population with Narrow Genetic Base M.Vales, J.Dossmann, J.Borrero

Population PCT-17 was delivered to the following associate countries:

Argentina University of La PlataChile National Agronomic Research Institute (INIA)China Agronomic Sciences Academy of Yunnan (YAAS)Colombia Rice Producers Federation (FEDEARROZ)Costa Rica University of Costa Rica (UCR)France CIRAD/ French Rice Center (CFR)Madagascar CIRAD/ National Center of Rural Development (FOFIFA).

This population is to develop varieties with resistance to cold of altitude orlatitude.

1.B.8. Adaptation to Manual Husk RemoverJ.I.Roa (PE3), H.Muñoz (Metálicas Metropolitana), M.Vales

After last year participative trials (Rice Project IP4 Annual Report 1999) a newmodel of a manual husk remover was constructed. Next year it will be tested inrural communities with small poor farmers.

1.B.9. Participation in Workshop, Seminar and CongressM.Vales

• Oral presentation "Upland Rice of Hillsides: An Option against FoodInsecurity", Monday April 24, 2000 in University of Costa Rica (UCR), CostaRica (Vales, 2000).

• Participation in Los SOLes workshop of CIAT’s Hillsides Program from April25 to 27, 2000, in San Dionisio, Nicaragua (Vales, 2000).

• Participation in the Third International Congress 2000 CROP SCIENCE, fromAugust 17 to 22, 2000, Hamburgo, Germany. (Vales et alii, 2000; Cf.summary in annex).

• Oral presentation "Upland rice for Hillside: an option against food insecurity inthe mountains", on September 1st, 2000, in the Food and Crop ResearchInstitute/ YAAS, Kunming, Yunnan, P. R. China.

• Oral presentation "Hillsides rice: an option against food insecurity in themountains" during CIRAD-CA Annual Meeting on September 5, 2000,Montpellier, France.

32

References

• Aventis Foundation. 2000. 22/06/2000 Aventis Foundation Press Release.The Fondation Aventis-Institut de France and CIRAD are joining forces tohelp develop sustainable agriculture in the countries of the SouthernHemisphere (Cf. Annexes).

• Surapong, S. 1991. A simplified crossing method for rice breeding. Amanual. CIAT : 32 p.

• Taillebois, J., and Castro. E. M. 1986. A new crossing technique. Int. RiceRes. Newsl. 11(3):6.

• Vales, M., Chatel , M.-H., Borrero, J., and Ospina, Y. 1998. RecurrentSelection for rice (Oryza sativa) blast (Magnaporthe grisea) Resistance inPopulation with Narrow Genetic Base. International Symposium on RiceGermplasm Evaluation and Enhancement, August 30 – September 2,Suttgart, Arkansas, U.S.A.

• Vales, M. 2000. Informe de viaje en América Central: Costa Rica -Nicaragua - El Salvador – Honduras. 23 de Abril, 5 de Mayo del 2000.Proyecto Colaborativo CIAT/CIRAD-CA, IP4 Arroz: 14 p.

• Vales, M, M.-C. Chatel, J. Borrero, E. Barrios, J.-I. Roa. 2000. Upland ricefor high altitude: an option against the food insecurity in the Andean hillsides.In: 3rd International Crop Science Congress 2000. 17-22 August 2000, CCH,Hamburg, Germany (Cf. Annexes).

33

Table 1.1. Progenitors for Highlands Rice

# Line Father Mother1 F4 22 CIRAD 391 / IRAT 265-57-2 Fertile, small grain2 45-8 IRAT 265-57-2 / Jumli Marshi Fertile, early, small grain3 F4 31 Jumli Marshi / Luluwini 22-M 2 Very black and long grain4 F4 18 Cuiabana / Miara 3 Long and fine grain5 F4 40 Miara / Latsibavy 2 Medium grain6 48-45 Pratao

Precoce/ Chhomrong Dhan Dwarf

7 F4 14 ChhomrongDhan

/ Slip 72-M-MA1 Fertile short medium grain

9 F4 25 IRAT 265-57-2 / Estrela Long and light yellow grain10 F4 13 Chhomrong

Dhan/ Miara Fertile, small grain

11 F4 17 Cuiabana / Long Sweet GlutinousRice

Long and fine grain, glutinous

12 F4 30 Jumli Marshi / Estrela Black small grain13 F4 38 Luluwini 22-M / Miara Long fine grain14 F4 39 Luluwini 22-M / Slip 72-M-MA1 Long and fine grain15 58-13 PRA 524 / PRA 630 Long grain16 58-1 PRA 524 / PRA 630 Long grain17 2-58 PRA 8 / IRAT 265-57-2 Medium grain18 F4 42 Shin Ei / Estrela Medium grain19 1-9 CA 148 / IREM 239 Medium grain

With:PRA 8 Latsidahy / IRAT 351PRA524

FOFIFA 60 / Chhomrong Dhan

PRA630

FOFIFA 116 / Chhomrong Dhan // Luluwini 1

CIRAD 391: the best variety in Popayan Montpellier: the best line for extra long fine grain

34

Table 1.2. Crosses for Hillsides Rice (Parents see table 1)

Mother Father1 / 151 / 52 / 42 / 112 / 132 / 152 / 166 / Montpellier7 / 147 / 102 / Montpellier9 / 3

12 / 317 / 15

CIRAD 391 / 4CIRAD 391 / 9CIRAD 391 / 13CIRAD 391 / 15CIRAD 391 / 17CIRAD 391 / Montpellier

Table 1.3. Parents for New Hillsides Rice Recurrent Population

# Mother Father Characteristics1 F4 22 CIRAD 391 / IRAT 265-57-2 Fertile, small grain2 45-8 IRAT 265-57-2 / Jumli Marshi Fertile, early, small grain3 F4 31 Jumli Marshi / Luluwini 22-M 2 Very black long grain4 F4 18 Cuiabana / Miara 3 Long fine grain5 F4 40 Miara / Latsibavy 2 Medium grain6 48-45 Pratao Precoce / Chhomrong Dhan Dwarf7 F4 14 Chhomrong Dhan / Slip 72-M-MA1 Fertile, short medium grain

35

Table 1.4. Ascending Parents (table 3) of New Hillsides Population

Ascending Characters being soughtIRAT 265-57-2 Partial durable Blast Resistance, adaptation to uplandJumli Marshi High resistance to cold (irrigated = dr)Chhomrong Dhan High resistance to cold, good parent for tilling (dr)Miara European Grain quality , high production (irrigated)Latsibavy Resistance sheath rot and cold (irrigated)CIRAD 391 The best in Popayan , Latsiday as ascendingCuiabana Very good parent for long fine grain, upland adaptationLuluwini 22-M Long fine grain of Cuiabana, upland adaptationPratao Precoce Upland adaptation, Blast resistanceSlip 72-M-MA1 Upland adaptation, Blast resistance

CIRAD 391 = Latsidahy / Shin Ei Latsidahy is a Latsika near LatsibavyLuluwini 22-M = Ciwini // Araguaia / CuiabanaIRAT 265 = IRAT 112 / IRAT 13 = IRAT 13 / Dourado Precoce // IRAT 13Slip 72-M-MA1 = IRAT 361 / IRAT 263 = Araguaia / Cuiabana // IRAT 112 / Iguape Cateto= Araguaia / Cuiabana /// IRAT 13 / Dourado Precoce // Iguape CatetoThus, in this last variety are Cuiabana and IRAT 112 already present in other lines,(In gray: cytoplasm)Sources of androesterility gene are individuals [ms] of a former recurrent population for upland rice forhillsides.

Table 1.5. Crosses for Hillsides Population (Parents see table 3 and 4)

Parents5 // [ms] / 14 // [ms] / 26 // [ms] / 37 // [ms] / 42 // [ms] / 51 // [ms] / 63 // [ms] / 7

Annex 1

22/06/2000. Aventis Foundation Press Release. Foundation Aventis-Institut ofFrance and CIRAD are joining forces to help develop sustainable agriculture incountries of the Southern Hemisphere.

(Paris, 22nd June 2000) -- The Fondation Aventis-Institut de France and CIRAD(Center for International Cooperation in Agronomic Research for Development)have signed a renewable three-year sponsorship agreement to allow developingcountries to share the benefits of sustainable agriculture.

36

True to its philosophy since it started in 1995, the Foundation seeks not just toprovide financial aid, but also to sponsor new initiatives and promote skills.Thus, teams from CIRAD, Aventis Crop Science and the Foundation set up theirfirst three programs. These are:Developing upland rice cultivation: first of all in the high plateau of Madagascar,and then in the Colombian Andes and in China.These regions are running the risk of exhausting their soils, due to a shortage ofarable land. Moreover, tropical soil is generally fragile and water resources forirrigation are limited.This program should provide solutions for cultivating rice on new soil, bybreeding new varieties and thus increasing the income of peoples living ataltitude, most of whom belong to impoverished ethnic minorities.Integrated Pest Management of sigatoka disease in citrus fruits: […]The third program, which overlaps the others will look at three crops (cotton inBenin, plantain in Cameroon and coffee in the Dominican Republic) […]

These first three programs meet the following criteria, which were jointlyestablished by the CIRAD and the Fondation Aventis-Institut de France:• Allow small farmers to raise their income, for example, by creating new

channels, micro-companies...,• Improve food security by increasing production of traditional crops and food,• Allow small farmers to choose methods which are more environmentally

friendly, by giving them alternatives to use plant protection products,• Try to involve everyone, bearing in mind local, social, organizational and

cultural differences, working mainly with village communities and other fieldplayers (NGO…).

Moreover, these three projects foster the transfer of experience and technologywith a view to sharing benefits of progress more fairly between the countries ofthe North and South. However, they also stress the importance of co-operationbetween the different countries and cultures of the South.Given how much is at stake, the joint program set up by the Fondation Aventis-Institut de France, Aventis Crop Science and CIRAD teams, is modest andpragmatic. Instead of just depending on technological progress alone, it willconsider the social, environmental and ethical aspects of each situation, and ofevery community involved in these programs, so that it may succeed, and so thattechnological progress can translate into real benefits for the countries in theSouthern hemisphere.

The Fondation Aventis (ex-Rhône-Poulenc), was created in 1995, under theauspices of the Institut de France. It provides financial support and skill sharingto projects which it selects with its partners, to encourage scientific innovation(assistance to young scientists setting up their own business…), to improvehealth and quality of life (research programs on orphan diseases...) andpromotes sustainable agricultural.

37

Annex 2

Summary. Upland Rice for High Altitude: An Option Against the FoodInsecurity in the Andean Hillsides

M. Vales, M.Chatel, J.Borrero1

Edmundo Barrios2, Jose Ignacio Roa3

1 Confronting food Insecurity in the Hillsides, a collaborative project between the Centrede Coopération Internationale en Recherche Agronomique pour le Développement(CIRAD) and Centro Internacional de Agronomía Tropical (CIAT); 2Confronting SoilDegradation of the CIAT Hillsides program; 3Participatory Agronomic Research programof CIAT; CIAT AA 6713 Cali, Colombia

Although rice consumption in the Andean hillsides is very high, it is usually notgrown in this region because of limited material tolerant to low temperatures.The consequence is the limited availability of rice in the diet of a considerableproportion of Andean families whose purchasing power is low.

The objectives of the project are:• Reduce food insecurity by the release of upland rice varieties with cold and

disease resistance.• Promote sustainable agricultural systems that incorporate adequate organic

crop management preventing erosion and agrochemical use, and thus protectthe soil and the water resources.

Since 1993, rice varieties were identified for potential release in the coffeegrowing zone of Colombia. Promising lines were identified in the Andean part ofthe Cauca department of Colombia, in breeding trials and participatory evaluationat the SOL (Spanish acronym (SA) for “Supermarket of Options for Hillsides”),and in 9 CIALs (SA for “local committees for agronomic investigation”). A narrowand a broad recurrent population were formed for a medium and long termbreeding process.

New funding is needed to complete the program in Colombia, and to enhance itin other Andean countries and Central America.

38


1.C. Advance and Evaluation of Inter-Specific PopulationsA. Almeida, J. Lopez, C. P. Martinez, J. Borrero, A. Bohorquez, G. Gallego, M. C.

Duque, J. Tohme, S. R. McCouch, P. Moncada, and M. Jones.

1.C.1. Introduction

Improving rice germplasm for Latin America and the Caribbean (LAC) by the IP-4Project at CIAT is being accomplished through the increase of genetic diversityand enhancement of gene pools for higher, more stable yields with adequatelevels of resistance to pathogen and insect pests relevant to the region. CIATbreeding strategies focus on developing and improving populations to provide allour partners with sources of potential parents having specific traits or populationsfrom which they may select and advance fixed lines for release as commercialvarieties.

In the case of irrigated and favored upland ecosystems, this is accomplishedthrough the characterization and utilization of wild rice species, development ofimproved populations through recurrent selection methods, and the introgressionof agronomic traits from the IRRI new plant type into our local gene pool.

1.C.2. Genes from Wild Rice Contribute to Yield Increase in Cultivated Rice

1.C.2.1. Introduction

Oryza wild species represent potential sources of new alleles for improving yield,quality and stress resistance of cultivated rice. Still, effective use of wild speciesgenes remains to unexplored. Advanced backcross-breeding schemes usingmolecular mapping techniques represent an alternative to reduce the geneticbackground from wild species parentals and to easily detect these alleles insegregating populations.

This collaborative project between CIAT, WARDA, and Cornell University, isaimed at characterizing and utilizing the genes from rice wild species for theimprovement of cultivated rice. Here we report progress made in theidentification of Quantitative Trait Loci associated with yield increase in Oryzarufipogon and Oryza barthii and the selection of families with these QTLs.

1.C.2.2. Materials and Methods

Experiments for the identification of segregant alleles from advancedbackcrosses were set up in the field, greenhouse and in the Biotechnologylaboratory of CIAT. BC2F2 segregant families (288) from cross Bg90-2/Orufipogon, and BC3F2 Segregant families from cross Lemont/O barthii (326) wereplanted in a randomized complete block design (RCBD) with two replications.

39

Planting was done on 4 row-plot of 5m length each, at CIAT- Palmira during June- December in 1996 and 1997. Data on 13 traits were taken on 10 randomlyselected plants. Based on yield potential and agronomic performance, 38 BC2F2families from Bg90-2/O.rufipogon cross were selected for grain yield evaluation,using a RCBD design with four replications.

Data on the13 agronomic traits were associated with the 127 molecular markersusing simple and multiple regression analysis (Nelson 1997). Proportion ofobserved phenotypic variance attributable to a particular QTL was estimated bythe coefficient of determination (R2) from corresponding linear model (singlepoint) analysis and multiple regression analysis.

DNA of young leaves from parental genotypes and segregating populations inboth crosses was extracted by the Dellaporta Method (McCouch, S. et al. 1988),modified for Polymerase Chain Reaction assay by CIAT Biotechnology ResearchUnit. A total of 127 markers (83 Random Fragment Length Polymorphism and44 SSRs) were used to evaluate the segregants from Bg90-2/O.rufipogon crosswhile the offspring of the Lemont/O.barthii cross were evaluated with 85 SSRsmarkers. These markers used in the evaluations and the QTL analyses wereselected from the rice molecular framework linkage map (10 - 20 cM intervalsthroughout the genome) (Causse, M. A. et al. 1994, Chen et al. 1997).

Based on the screening of 210 SSR in parents of both crosses (Temnykh, S. etal. 2000), the PCR assay protocols for 40 of them have been standardized foruse in the screening of the progeny the Lemont/O.barthii cross.

From the results obtained in the molecular analyses of the 288 BC2F2 familiesfrom the Bg90-2/O.rufipogon cross, 87 families that were associated with QTLsfrom O.rufipogon, were selected to start development of NILs. A total of 40plants from each family were planted in the greenhouse and later transplanted inthe field. Data on 13 agronomic traits were taken and used to select morepromising families for the traits of interest.

While the plants were still in the greenhouse, leaf discs (diameter of 5mm) weretaken from each plant of the selected BC2F2 families for DNA extraction usingthe Alkali Method (Klimyuk, V. I. et al. 1993). Approximately 40 SSRs wereanalyzed in a total of 235 PCR assays done with the 87 families. Plants carryinghomozygote or heterozygote alleles for O.rufipogon genotype were chosen andbackcrossed to Bg90-2. BC3F1 seed was obtained to continue the developmentof NILs carrying specific QTLs. Several cross combinations were made tocombine different QTLs derived from O.rufipogon.

1.C.2.3. Results and Discussion

Based on the 83 RFLP and 44 SSR from the RF- Cornell framework mapscreened on 288 BC2F2 families from the cross Bg90-2/O rufipogon, putative

40

linkages were identified with yield, and yield components from replicated dataavailable for the whole mapping population. It should be considered that themultiple regression analysis we used associates quantitative traits (yi) as afunction of xi (number of alleles A per locus or marker). Homozygous genotypeslike Bg90-2 will have xi =2 and heterozygous ones will have xi =1. The signassociated with the coefficient indicates which parent is contributing the alleleresponsible for a higher phenotypic mean. A negative sign indicates that ahigher value is due to O.rufipogon. The magnitude of the coefficient shows therelative contribution of the putative QTL on a particular trait. It is a combinedfunction resulting from the interaction between the recombination value andadditivity.

Results using the Qgene software for molecular breeding, version 3.0 (Nelson2000) indicate associations between markers and yield on chromosomes 3, 5, 6,9 y 12. Results obtained in chromosomes 5 and 12 showed similar associationsto that reported by Xiao et al. 1996 and 1998 (Fig. 1 and table 1). Duncan's testswere done to compare mean yield of F2, F3, and F5 generations. Statisticalanalysis (not shown) showed that ,except for two F2 families, there was no yielddifference among generations, which suggest that hybrid vigor was not involved,and that yield differences observed in the F2 generation were maintained throughseveral generations of phenotypic selection. Combined analysis of yield data(not shown) also showed that lines CT13976-7-M-6-M, CT13956-29-M-3-M, andCT13941-27-M-14-M had a mean yield significantly higher than Bg90-2 atP=0.05. Molecular characterization of these lines will determine which QTLs areassociated with the increase in grain yield. Some inter-specific lines were sent toNARs in LAC this year for evaluations under local conditions.

41

Table 1.6. Statistical Analyses of BC2F2 Population, Multiple Regression forYield and Plant Height Traits, Threshold Value 5% (Qgene, Nelson 2000)

Quantitative Chr Markers Ajusted F Coeff. T P Increment Add%Trait -Sq Total (2 Colas) EffectYield Intercept 0.164 8.2 2166.325 2.46 0.0146

3 RG100 403.2821 2.848 0.0048 Bg90 12.845 RM13 -317.629 -2.047 0.0418 rufipogon -13.396 RM217 R 742.3548 2.886 0.0043 Bg90 13.019 RM215 -404.861 -2.686 0.0078 rufipogon -13.34

12 RG901 630.5277 2.871 0.0045 Bg90 12.9212 G1112 441.4269 2.506 0.0129 Bg90 12.9

Plant Intercept 0.746 137.84 152.313 47.03 0Height 1 RZ538 -22.1423 -24.603 0 rufipogon -48.09

2 RM233A -3.4199 -4.695 0 rufipogon -6.766 RM3 -4.2506 -3.577 0.0004 rufipogon -10.18 RM38 -1.9829 -2.267 0.0243 rufipogon -10.83

12 RG341 1.78 2.411 0.0167 rufipogon -5.88

In addition, 61 other QTL associations with yield and various yield componentswere detected. For example, five markers were found associated with plantheight on chromosomes 1,2,6,8, and 12 RZ538 explained 70 percent of thephenotypic variation in plant height in this BC2F2 cross (Fig.1.1 and table 1.6),and showed a similar location on the chromosome as reported by Xiong etal.1999. These results suggest that each QTL derived from O.rufipogon has adifferent effect on plant height, which gives breeders the opportunity to developimproved breeding lines having different plant height to suit diverse growingconditions.

Preliminary results have gotten the attention of NARs, which are requesting tospeed up the development of improved inter-specific lines to be used as potentialparents in their breeding program. To this end 74 inter-specific crosses weremade this year to transfer useful traits from wild rice into improved cultivars. Inaddition, several populations (F1, F2, F3, and F5) were planted in pedigree rows(3036) in CIAT-Palmira and Santa Rosa, Villavicencio. Approximately, 7064 and233 plant selections were made in Palmira and Santa Rosa, respectively forfurther evaluations in 2001.

1.C.3. Research Highlights with Collaborating Institutions

1.C.3.1. WARDA

In 1997 WARDA initiated the Africa/Asia joint research project on interspecifichybridization between the African and Asian rice species for the development ofimproved rice varieties having higher weed competitiveness, resistant to the

42

African rice gall midge, drought, iron toxicity, good grain quality. CIAT isparticipating in the development of improved breeding populations throughinterspecific crosses and evaluation of breeding lines.

Seed of 100 interspecific O.sativa/O.glaberrima progenies were received fromWARDA in 1998 and planted in observational plots under upland acid soilconditions in La Libertad Experimental Station, Villavicencio. Oryzica Sabana 6and O. sabana 10, and Linea 30 were planted as control checks for comparison.Nearly 60% of the lines were susceptible to rice blast but most of the linesshowed good tolerance to leaf scald and Helminthosporium. Based onpreliminary data on field performance and yield potential 36 lines were selectedfor further evaluation in 1999. These selected progenies continued to show highresistance to main diseases such as leaf and neck blast (P. oryzae), brown spot(H.oryzae), and grain discoloration, as well as tolerance to acidic soil conditions.Some of them had better seedling vigor and earliness than local checks. LinesWAB450-1-B-P-82-2-1, WAB450-1B-P-91-HB, WAB450-1-B-P-133-HB,WAB450-1-B-P-6-2-1, and WAB450-1-B-P-92-3-1 yielded as well as local checks(2 ton/ha). However, in terms of grain quality these interspecific progenies do notmeet consumer preferences in Latin America for long and slender, translucentgrain type. Some of these lines are being used as progenitors in our breedingprogram. WARDA in Sept. during a visit to 2000, 300 interspecific lines wereselected from WARDA's nurseries for evaluation in Villavicencio in 2001.

305 lines derived from the BC2F2 generation of Bg90-2 and O.rufipogon weresent to WARDA in 1998 for evaluation under the different ecologies found inWest Africa. Many promising lines were identified and selected by Warda'sBreeders. During this year, nearly 1283 segregating breeding lines derived frominterspecific crosses were sent to WARDA for evaluation.

1.C.3.2. Cornell University

A pilot project on rice was initiated between CIAT and Cornell in 1994. Basis ofthis project is a forward- looking strategy for improving crop performance usingexisting collections of genetic resources and the tools of biotechnology. Crossesinvolving Oryza glaberrima, Oryza rufipogon, and Oryza barthii in combinationwith elite irrigated and upland varieties were made and several improvedpopulations have been developed so far. An advanced backcross breedingstrategy was used to identify quantitative trait loci (QTL) associated with eightagronomic traits in a BC2F2 population derived from a cross between Caiapo, anupland Oryza sativa subsp. Japonica rice from Brazil, and an accession of Oryzarufipogon from Malaysia. Based on analysis of 125 SSLP and RFLP markersdistributed throughout the genome and using single point, interval, andcomposite interval mapping, two putative O.rufipogon derived QTL were detectedfor yield, 13 for yield components, four for maturity and six for plant height. Itwas concluded that advanced backcross QTL analysis offers a useful germplasmenhancement strategy for the genetic improvement of cultivars adapted to stress

43

prone environments. Although the phenotypic performance of the wildgermplasm would no suggest its value as a breeding parent, it is noteworthy that51% of the trait enhancing QTLs identified in this study were derived fromO.rufipogon. A paper (Moncada et al, 2000) was submitted and accepted forpublication by Theoretical Applied Genetics.

1.C.4. Population Improvement Using Gene Pools and Population withRecessive Male-Sterile Gene

J. Borrero, J.Carabalí, C.Martinez


Recurrent selection is broadly defined as the systematic selection of desirableindividuals from a population followed by recombination of the selectedindividuals to form a new population. The process is envisioned as a circle thatincludes population development, evaluations of individuals, and selection ofsuperior individuals as parents to form a new population for the next cycle ofselection (Fehr,1987). It is a dynamic and continuous process aimed atdeveloping superior genotypes in one or more traits. Since 1989 CIAT riceproject is using recurrent selection methods as one of the strategies to increasethe yield potential and resistance to biotic and abiotic production constraints.This activity aims at: a) developing improved populations for irrigated and favoredupland conditions having a higher yield potential and good grain quality; b)provide regional partners with parental sources and / or lines with specific traits;c) promote networking among interested partners.

1.C.4.2. Materials and Results

Four base populations (PCT6,PCT7, PCT8 and GPCT9) developed earlier wereplanted and evaluated in CIAT-Palmira in terms of yield potential, grow duration,early vigor, plant type, and grain quality. Based on our observations andfeedback from regional partners PCT6, and PCT8 were selected for furtherimprovement in yield potential and grain quality. Based in yield potential andfield performance, 14 cultivars were selected and crossed to male-sterile plantsfrom each base population to generate more variability. In addition, 13 cultivarsknown to have excellent grain quality were also selected and crossed to severalmale-sterile plants from each selected base population. In all four, newpopulations were formed for further improvement and selection in 2001.

44


1.D. Introgression of New Plant Type Genes Into LAC's Gene PoolsZ. Lentini, A.Mora, J.Borrero, C.Bruzzone, C.Martinez

1.D.1. Introduction

Ideotype breeding aimed at modifying the plant architecture is a time testedstrategy to achieve increases in yield potential. To attain additional increases inrice yield potential, IRRI scientists conceptualized a new plant type (NTP) in1988. Further modifications in plant architecture were proposed with thefollowing characteristics: low tillering capacity, no unproductive tillers, 200 to 250grains per panicle, very sturdy stems, dark green, thick erect leaves, andvigorous / deep root system. Numerous breeding lines with desired ideotypewere introduced to CIAT and evaluated in Palmira and Santa Rosa foradaptation, yield potential, grain quality, and tolerance to main biotic and abioticstresses. Several promising lines were identified and used as parents in ourbreeding program. This activity is aimed at incorporating important agronomictraits exhibited by the NPT material into LAC's gene pools.

1.D.2. Materials and Methods

A dual strategy was followed. Two hundred ninety nine F1 populations werederived from crosses having at least one dosage of a NPT parent and twodosages of a locally adapted cultivar. Some populations (104) were run throughanther culture to speed up the development of fixed homozygous lines and theremaining ones were sown in pedigree rows. Out of (dh) approximately 1000 R0green plants regenerated last year 318 doubled-haploid plants were selected andevaluated in Palmira and Santa Rosa. Based in yield potential and tolerance tomain diseases 64 Dhs lines were selected for distribution to NARs in 2001.Out of 43 F1 populations derived from crosses between a NPT parent and LAC'smaterial 365 single plant selections were made in Palmira. In addition, 611 F3lines were evaluated in pedigree rows in Palmira. Good variability in terms ofplant and grain types, maturity duration, tillering ability, panicle size, strawstrength, number of grain per panicle and plant height was observed. A total of1381 plant selections were made for further testing in 2001.

1.D.3. Training ActivitiesA.Caldas, C.Martinez, J.Borrero, E.Garcia, D.Salgado

An international course on application of conventional breeding methods andmolecular techniques for the genetic improvement of rice was organized in CIATPalmira. A total of 22 rice scientists from 6 countries (Colombia, Costa Rica,Brazil, Dominican Republic, Uruguay and Venezuela) participated in a two-weekcourse aimed at integrating the use of conventional methods and molecular toolsin rice breeding. Participants felt very strongly that courses like this one are ingreat demand and should be offered on a regular basis.

45

CHR3

RM148

RZ630

RM55

RM16

RG369

RG913

RM251

RM7

RG100

RZ545

RM22

RM60

GL, PL,

YLD, WHC

GL

WHC

CHR4RG169BCD135RZ590RG214RG476RG329RZ717RG177RZ656RG908RZ69RG375RM261

GPP, PL, GRPPL

G

GPL

CHR6

RZ884

RZ682

RG1028

RM3

RZ612

RG445

RZ667

RM217

RM204

RZ1002

YLD, DTH

PL, H

GL

CHR1

RZ730

RM212

RZ444

RZ538

RG957

RM5

RZ995

RM1

RM220

RZ543

GPP, PL, GPL, PS,DTH

GRPP

GPP, PS

CHR2RM48

RM208

CDO686

RM221

RZ668

RG25

RM 6

RZ476

RZ599

RG509

RM233A

CDO524

PPL

GPL

GPP

H

CHR5

RM31

RM26

RZ225

RG474

CDO1160

RZ67

RM164

RM249

RM13

RM122

GW, GPP, WHC

GL, WHC

PL, GPL

WHC

GL

YLD

CHR7

RZ272

RG30

CDO407

RM214

RM11

RM234

RM18

RG146

RM248

CHR

RM38

RZ323

RM44

RZ617

RG1034

RM210

RM230

RG598

PPL, DTH

DTH, H

CHR

RM205

RZ404

RG451

RM215

RM242

RM257

YLD, PL,

PPL,

CHR10RZ421

RZ811

BCD386

RM216

RM222

PS

CHR1RZ536RM224RG303RM254C82G1465RM21G44G320BCD808RM167C794RM20BRM4

GL

GL

PPL, GPLGW

CHR1

RM17

RG901

RG190

RG543

RZ76

RZ261

RG457

RG9

RG869

RG81

RG341

G1391

RZ397

RM19

RZ816

G1112

RG574

RM20A

YLD,

YLD

PL

GPL

Figure 1. Molecular mapping and QTL alleles foryield and components (Bg90-2/O.rufipogon.Multiple regressions, threshold value 5% and 1%(Bold and Italic traits). The quantitative traitsassociated with yield increase due to O.rufipogonare underlined.

46

On-Going Activities

• Complete characterization of agronomic and molecular data, and QTLanalyses to determinate the number of QTLs associated with yield increaseacross environments for Bg90-2/O. rufipogon and Lemont /O. barthii crosses.

• Develop NILs carrying specific QTLs for use in breeding programs from thecross Bg90-2/O. rufipogon.

• Develop mapping population (CG14/ WAB56-104) in collaboration withWarda.

• Fingerprinting of advanced lines from Bg90-2/O.rufipogon and Lemont/O.barthii crosses.

• Initiate agronomic and molecular characterization of several other populationsinvolving crosses with O.glaberrima to identify QTLs associated with yieldincrease.

References

Causse, M. A.; Fulton, T. M.; Cho, Y. G.; Nag Ahn, S.; Chunwongse, J.; Wu, K.;Xiao, J.; Yu, Z.; Ronald , P.; Harrington, S. E.; Second, G.; McCouch, S. R. andTanksley, S. D. 1994. Saturated molecular map of the rice genome based on aninterspecific backcross population. Genetics, 138: 1251- 1274.

Chen, X.; Temnykh, S.; Xu,Y.; Cho, Y. G. and McCouch, S. 1997. Developmentof a microsatellite framework map providing genome-wide coverage in rice(Oryza sativa L.). Theor. Appl. Genet. 95: 553- 567.

Klimyuk, V. I.; Carroll, B. J.; Thomas, C. M. and Jones, J. D. G. 1993. Alkalitreatment for rapid preparation of plant material for reliable PCR analysis. PlantJournal. 3,3: 493- 494.

McCouch, S.; Kochert, G.; Yu, Z.; Wang, Z.; Khush, G.; Coffman, W. andTanskley, S. 1988. Molecular mapping in rice chromosomes. Theor. Appl.Genet. 76: 815- 829.

Temnykh, S.; Park, W. D.; Ayres, N.; Cartinhour, S.; Hauck, N.; Lipovich, L.; Cho,Y. G.; Ishii, T. and McCouch, S. 2000. Mapping and genome organization ofmicrosatellite sequences in rice (O. sativa L.). Theor. Appl. Genet.100: 697-712.

Xiao, J.; Grandillo, S.; Ahn, S.; McCouch, S. and Tanskley, S. 1996. Genesfrom wild rice improve yield. Nature. 384: 223- 224.

47

Xiao, J.; Li, J.; Yuan, L. and Tanksley, S. 1996. Identification of QTLs affectingtraits of agronomic importance in a recombinant inbred population derived from asubspecific rice cross. Theor. Appl. Genet. 92: 230- 244.Xiao, J.; Li, J.; Grandillo, S.; Nag Anh, S.; Yuan, L.; Tanksley, S. and McCouch,S. 1998. Identification of trait- improving quantitative trait loci alleles from a wildrice relative, Oryza rufipogon. Genetics. 150: 899-909.

Xiong, L. Z.; Liu, K. D.; Dai, X. K.; Xu, C. G. and Zhang Q. 1999. Identification ofgenetic factors controlling domestication- related traits of rice using an F2population of a cross between Oryza sativa and O. rufipogon. Theor.Appl.Genet. 98: 243- 251.

1.D.4. Publications

• Moncada, P.; C. P. Martinez; J. Borrero, M. Chatel; H. Gauch Jr.; E.Guimaraes; J. Tohme, and S. R. McCouch. 2000. Quantitative trait loci foryield and yield components in an Oryza sativa/Oryza glaberrima BC2F2population evaluated in an upland environment. Theor. Appl. Genet (inpress).

1.D.5. Conferences

• 28th Rice Technical Working Group. February 27-March 1.2000. Biloxi,Mississippi. A paper entitled “Advanced backcross analysis for the Transferof QTLs from O.rufipogon and O.barthii " was presented.

• Chandler Memorial Symposium. June 15-17. Cornell University. Ithaca. NY.

• Visit to Yale University at New Heaven, University of Minnesota at St Paul,and Novartis. June 20-28.

• Visit to Warda. Sept. 10-18.Bouake, Ivory Coast. Coordination ofCIAT/Warda collaborative activities in rice.

• 4th International Symposium on Rice Genetics. October 23-27, IRRI, LosBaños, Phillippines. A paper entitled "Utilization of new alleles from the wildrice Oryza rufipogon to improve cultivated rice (Oryza sativa) in LatinAmerica" will be presented.

48


1.E. The Use of Anther Culture and In Vitro Culture for Enhancement ofGene Pools

A. Mora (IP4),G.Delgado (IP4), T.Agrono (IP4), C. Ordoñez (IP4),J.Borrero (IP4), C. Martínez (IP4, SB2), Lentini (SB2, IP4).

1.E.1. Introduction

The CIAT rice anther culture laboratory (ACL) focus on developing doubledhaploid lines for the various breeding efforts. In the case of CIAT, the work hasbeen mainly directed to advanced populations adapted to the irrigated andupland savanna ecosystems, as well as back-crossed populations derived frominter-specific crosses between cultivated rice and wild rice species. In the caseof FLAR. The laboratory has generated doubled haploid lines from FLARcrosses targeting the sub-tropical and cold tolerant breeding lines for theSouthern cone. From September 1999 to September 2000, the ACL processed335 crosses generating a total of 5,499 plants. This year the level of activity inthe ACL was reduced in about 50% respect to the period 1998-1999 when a totalof 11,655 plants had been produced. This change in the level of activity for thisyear corresponds to a non-anticipated reduction from 364 crosses originallyrequested by FLAR for the work plan-2000, to a total of 190 crosses finallyprocessed. To balance this reduction in activity, the ACL staff and correspondingoperational funds gave for the first time support to the rice genetic transformationresearch activities, which had been supported by the SB2 project.

1.E.2. Use of Anther Culture to Fix Enhanced Traits in Back-CrossedPopulations from Rice X Wild Species Hybrids

A. Mora (IP4),G.Delgado (IP4), T.Agrono (IP4), C. Ordoñez (IP4),J.Borrero (IP4), C. Martínez (IP4, SB2), Lentini (SB2, IP4).

• Last year it was reported the generation of 3,309 green plants from the crossbetween the variety Progresso and the wild species O.barthii. R3 derived linesare currently being evaluated for yield potential.

• A total of 354 R3 lines were selected from the cross Lemont/O.barthii• This year a total of 99 selected BC2F2 and BC3F1 lines from the cross Caiapo⁄

O.glaberrima were processed through anther culture.• Anther culture is being used to recover fertile advanced lines, since most of

the back-crosses showed high level of sterility.• Doubled haploids lines will be selected and evaluated for disease resistance

and agronomic performance, and analyzed by molecular markers.• The main objective of this activity is to use doubled haploids for accelerating

the introgression of QTLs associated with high yield potential from the wildspecies into the selected O.sativa Caiapo variety.

49

• The introgression of an increased drought tolerance and weed competitioncapacity from O.glaberrima into this variety adapted to upland and acid soilconditions will also be evaluated.

1.E.3. Use of Anther Culture to Advanced Breeding Populations for TropicalIrrigated Conditions

A. Mora (IP4),G.Delgado (IP4), T. Agrono (IP4) C. Ordoñez (IP4),J. Borrero (IP4), C. Bruzzone (IP4), C. Martínez (IP4), Z. Lentini (IP4).

• A total of 318 R2 generated from corsses with the new plant type wereselected in Palmira for plant type and yield potential. Of these, 63 R2 linesalso combined disease resistance selected from Santa Rosa ExperimentalStation.

• Forty F1 crosses directed to tropical and subtropical conditions were selectedfor processing through anther culture generating 691 plants.

• Seed increase from R2 lines was obtained, and evaluation for diseaseresistance was conducted at Santa Rosa experimental station.

• Sister R2 lines are being evaluated for plant type, and some components ofyield such as tillering capacity, days to flowering, panicle length, fertility, andgrain size at CIAT experimental station.

• Selected R3 lines will be evaluated for yield potential next season, andadvanced R4 distributed to national programs

1.E.4. Use of Anther Culture to Fix Cold Tolerance in Recurrent SelectionPopulations

A. Mora (IP4),G.Delgado (IP4), T. Agrono (IP4) C. Ordoñez (IP4),J. Borrero (IP4), M. Chatel (IP4), Z. Lentini (IP4).

• Two different recurrent selection populations carrying male fertility restorergene along with cold tolerance were cultured.

• One thousand two hundred ninety eight plants were produced.• Total of 325 R2 plants were selected for disease resistance and sent to Chile

to be planted this November 2000 for final field evaluations.• Four crosses combining cold tolerance, high yield and good quality traits were

processed through anther culture• Six hundred and ninety one plants were generated, and R2 are currently being

harvested

1.E.5. Use of Anther Culture (AC) to Accelerate the Development ofBreeding Populations of FLAR.

A.Mora (IP4),G.Delgado (IP4), T. Agrono (IP4) C. Ordoñez (IP4),L.E. Berrio (FLAR), J. Gibbons (FLAR1), Z.Lentini (IP4).

1 Current address: Nebraska Rice Research Center, USA

50

• Total of 190 triple crosses designed for Southern Brazil were cultured fromOctober 1999 to February 2000.

• Plants per each line were first selected using as reference checks selectedBrazilian varieties. The selection was based on plant type, days to floweringand grain length (as predicted by the floret size), and the anthers from all theselected plants were bulked per cross and cultured.

• The 190 crosses yielded a total of 1,554 plants.• R2 plants from the R1 doubled haploid plants will be generated from these

crosses and will be selected both at Palmira Station for plant and grain type,and at Santa Rosa Experimental Station for blast resistance.

• A total of 109 R2 and 21 R3 lines generated from crosses processed in 1999,were selected for disease resistance at Santa Rosa and for plant type andgrain quality at CIAT headquarters, and included in the VIOFLAR for theSouthern Cone this year. The 130 doubled haploid lines represents __% ofthe lines sent in the VIOFLAR 2000.

1.E.6. Somaclonal Variation to Increase Genetic Variability of AdvancedBreeding Lines of FLAR Member Countries.

A.Mora (IP4),G.Delgado (IP4), T. Agrono (IP4) C. Ordoñez (IP4), Edgar Torres(Fonaiap, Venezuela), J. Holguín (FEDEARROZ, Colombia), J. Gibbons (FLAR1),

Z.Lentini (IP4).

• Responding to a request from FLAR research group at CIAT headquarters,and Colombia and Venezuela as members of FLAR, the ACL had generatedsomaclone lines derived from immature inflorescence using selectedvarieties. The goal of this activity is to induce variation for improving grainquality traits, RHBV resistance, tolerance to sogata mechanical damage, andlodging tolerance.

• CIAT had discouraged FLAR to use this approach for generating variants dueto its low efficacy; however, CIAT agreed to generate the lines requested byFLAR.

• Last year a total of 3,309 somaclones were produced for the VenezuelanNational Plan of Rice leaded by Fundarroz, and 4,440 somaclone plants weregenerated for FEDEARROZ, Colombia.

• Preliminary evaluation on S1 plants showed some somaclones with improvedgrain quality or resistance to RHBV. However, these traits were not inheritedin the S2 plants indicating the presence of epigenetic variation commonlyfound in this type of in vitro culture derived plants.

• This year 3,178 somaclones were generated from two varieties forFUNDARROZ-Venezuela.

• The S1 seed (first self of the original somaclone, S0) will be harvested, and S1plants will evaluated for grain quality. Disease resistance will be evaluated onthe S2.

• Because of the low efficiency of the procedure, CIAT decided not to continuewith this activity.

51

OUTPUT 2. CHARACTERIZING RICE PESTS AND THE GENETICS OFRESISTANCE

2.A. Characterization and Genetics of Resistance to Rice Blast, SheathBlight and Grain Discoloration

F.Correa

2.A.1. Introduction

Rice diseases are major production constraints in Latin America. These diseasescause severe losses in the vegetative and reproductive stages of the crop.Development of resistant cultivars has been the preferred means of controllingthese diseases, however, development of resistant cultivars with durableresistance has been a very difficult task, especially in upland areas of the LatinAmerican tropics where the climatic conditions are highly conducive to them. Ingeneral, resistance to pathogens breaksdown shortly after cultivar release.Nonetheless, host plant resistance has been, and continues to be, the preferredmeans of dealing with pathogens for a number of reasons. Major efforts arebeing made at CIAT to understand the high pathogen variation observed, oftenreported as the main cause of resistance breakdown. Extensive studies ofpathogen diversity have been initiated by analyzing the population structure indifferent Latin American countries.

Outputs

2.A.2. Characterization of Blast Pathogen Populations. Monitoring theEvolution in the Genetic and Virulence Diversity of the Blast PathogenOvertime (Project in collaboration with CIRAD)

F.Correa, D.Tharreau, E.Tulande, F.Escobar, G.Prado, G.Aricapa

The current series of experiments aim at following blast population genetic andpathogenic (virulence and agressiveness) changes on varieties generatingselection pressure by mean of complete or partial resistance, and susceptibility tothe pathogen. In the field, the resistance level of the varieties studied is alsobeing surveyed by mean of disease evaluation. Relations between blastpopulation changes and evolution of disease level will be examined for a periodof three years.

One experiment is being conducted at Santa Rosa with the following varieties:Complete resistance: Fedearroz 50, Oryzica Llanos 5; Partial resistance:Ceysvoni, Oryzica Llanos 4, IR 36, Iniap 11 (IR 64), Oryzica 2; Susceptiblevarieties: Oryzica 1, Oryzica Caribe 8, Cica 8, Oryzica 3. Each variety is beingsown every 15 days 6 dates in plots 5x5 meters. Evaluation of disease level inthe field is realized 30, 37, 44, 51 and 58 days after sowing. Fifty diseasedleaves are collected 37 days after sowing for each variety and date of sowing.Ten diseased panicles are collected for each variety and date of sowing 25 days

52

after flowering. Genetic analysis and pathotyping is being carried out on the blastsamples using PCR-DNA fingerprinting and greenhouse inoculations.

DNA samples from more than 100 isolates of Magnaporthe grisea collected inthe first planting date in 1999 were fingerprinted in 2000. This was done by usingrepetitive element-based polymerase chain reaction (rep-PCR) with twooutwardly directed primer sequences from Pot 2, an element found inapproximately 100 copies in the blast fungal genome (Goerge et al.). All isolatesanalyzed so far (116) were clustered in three major groups corresponding to thewell known Colombian genetic lineages SRL-6B, SRL-6, and SRL-4 (Figure 2.1,table 2.1, table 2.2). Differences in the genetic structure, as suggested in Figure2.1, are being analyzed in replicated experiments in order to be associated toevolutionary processes and response to the selection pressure of the ricecultivars with different resistance levels. Spectrum of virulence of 34 isolates ofthese 3 lineages representing most of the genotypic diversity observed in Figure2.1 was determined in greenhouse studies and shown in table 2.1. No singleresistance gene is effective against all isolates of the three detected lineages butsome of the known genes are effective against all isolates of the individualgenetic lineages studied. The near isogenic line C 101 LAC is the only lineexhibiting resistance against the three lineages, however our studies indicate thatthis line has the two distinct resistance genes Pi-1 and Pi-11. This isoline hasbeen reported in the past as carrying only the gene Pi-1, however all isolates inlineage SRL-6 infected the isolines C 104 LAC and C 103 TTP carrying theresistance gene Pi-1 but not the isoline C 101 LAC. The presence of theresistance gene Pi-11 in C 101 LAC was corroborated in a study conducted byus at CIRAD and will be discussed later in this report. These two resistancegenes are defeated however by isolates of lineage SRL-5 included in table 1 forreference. When in combination may provide resistance to all isolates of this andsee the Colombian lineages. The three resistance genes Pi-1, Pi- 11, and Pi-2opening up one of the few possibilities for developing a true resistantcombination of genes to the whole pathogen population in Colombia. It seemsthat the gene Pi-1 confers resistance to the lineages SRL 6B and SRL-4, thegene Pi-11 to SRL 6, and the gene Pi-2 to lineage SRL-5. We expect to have inthe near future molecular markers associated with these three genes in order tofacilitate their incorporation in different rice cultivars.

The most common lineage recovered from infected leaves so far has been SRL-4 (table 2.2), which has been present in all five cultivars sampled up to now. Thethree lineages (SRL6B, SRL6, and SRL-4) have been recovered from thecultivars Oryzica 1 and Ceysvoni while the cultivars Oryzica Caribe 8, OryzicaLlanos 5 and IR 36 have yielded only isolates in lineage SRL-4 (table 2.2). Not allisolates recovered from lineage SRL-4, and specifically those recovered from thecultivars Ceysvoni and Oryzica Llanos 5 have reinfected the cultivar of origin inthe greeenhouse inoculations (table 2.3). These two cultivars exhibit a partialresistant and complete resistant reaction, respectively, while those cultivars(Oryzica 1 and Oryzica Caribe 8) reinfected by isolates recovered from them

53

exhibit a highly susceptible reaction. We are presently analyzing more isolatesfrom the partial and resistant cultivars in order to understand the meaning of ourresults in evolutionary terms of the pathogen and its possible relationship withthe breakdown of the resistance.

Recovery of blast isolates from the highly resistant cultivars Oryzica Llanos 5 andFedearroz 50 has been difficult due to the low number of lesions on the leaves.Fedearroz 50 exhibited a low incidence of neck blast at the first planting date of1999 and 23 isolates were recovered from the infected panicles for pathogenicitytests. Different sources of Fedearroz 50 were used in the inoculations toeliminate the possibility of seed mixtures in the field. Three of these isolatesreinfected the cultivar Fedearroz 50 in the inoculation tests on the leaves as wellas other cultivars tested (table 2.4) suggesting the potential breakdown of theresistance of this cultivar. These isolates did not infect the near isogenic line C101 LAC indicating that the resistance genes Pi-1 and Pi-11 are effective inconferring resistance to this new variant of the pathogen. We are currentlyanalyzing the genetic structure as well as the virulence composition of thesecompatible isolates to determine the changes occurring in this new populationand how it is associated to the selection pressure exerted by rice cultivars withhigh levels of resistance. None of the 33 isolates tested for pathogenicity (table2.1) showed compatibility with the cultivar Fedearroz 50 when these isolateswere inoculated on all the Colombian rice varieties (data not shown).

We have introduced SCAR (sequence characterized amplified regions) markers (10 pairs of primers) developed by CIRAD in order to complement and facilitatethe analysis of the Magnaporthe grisea populations considered in these studies.These markers have been developed from RAPD markers linked to avirulencegenes as well as from different published sequences of other M. grisea genes.The main advantage of this technique is that it generates simple profiles that canbe easily scored and compared while the rep-PCR and MGR techniques aremore difficult to score due to the number of bands. We expect to transfer thistechnique to many laboratories in the region as soon as we test the suitability ofthe SCAR markers.

2.A.3. Selection of Rice Blast Resistance Sources to Different Blast GeneticLineages under Greenhouse and Field Conditions

F.Correa, G.Prado, G.Aricapa, E.Tulande, F.Escobar

The frequency of blast resistant plants observed in F2 populations in the field ishighly dependent on the blast reaction and stability of this reaction of the parentsused for the development of these populations. An increase in the number of F2susceptible plants in 2000 compared to previous years has been observed andwas related to the low stability of the blast resistance of the parents used in thebreeding program. We have initiated the blast evaluation in the field andgreenhouse of several hundred advanced lines exhibiting desired agronomictraits for the identification of blast resistance sources. The selected resistant lineswill be tested in several seasons and inoculation tests with blast isolates

54

representing most of the virulence and genetic diversity of the pathogen todetermine the stability of their resistant reaction before being recommended as aparent. Blast lesions observed in low frequency in the field on these cultivars willbe collected for isolate recovering and greenhouse inoculations. We aredeveloping a nursery of potential sources of resistance, which will be used asparents after several periods of testing. A group of rice lines exhibiting a highlyresistant leaf and panicle blast reaction at Santa Rosa under high diseasepressure in two replications were selected in 2000 and included in this nursery(table 2.5). Additionally, another group of advanced lines from FLAR and CIATwere selected for their resistance to the six Colombian genetic lineages ingreenhouse inoculations and field resistance at Santa Rosa (table 2.6) andincluded also in the nursery. These lines had also exhibited a resistant blastreaction at Santa Rosa in 1999.

2.A.4. Characterization of the Genetic Resistance to Rhizoctonia solani(sheath blight). Development of Inoculation and Evaluation Methods

F.Correa, G.Prado, G.Aricapa, and F.Escobar

The filamentous basidiomycete Thanatephorus cucumeris (anamorph= R.solani)is the causal agent of sheath blight of rice. This disease has increased ineconomic importance in most Asian countries as well as in the USA in the last 10years. The disease is also increasing in importance in most rice growingcountries of tropical Latin America where the specie R. solani AG-1 IA seems tobe the most common while the specie R. oryza-sativae seems to be the mostcommon in the temperate areas of South America. The disease, which hasincreased in incidence and severity, is most frequently controlled with the use offungicides. There are not well known sources of resistance to the pathogen. Wehave initiated the evaluation in the greenhouse of different rice lines in order toidentify potential sources of resistance to this pathogen. The evaluatedgermplasm includes Colombian commercial cultivars, wild rice species, Asianand USA reported sources of resistance, and advanced breeding lines of the riceproject.

Greenhouse inoculations were performed using 14 different sheath blight isolatescollected in different areas of Colombia since 1988. Most Colombian commercialcultivars tested exhibited and intermediate or susceptible sheath blight reaction(table 2.7). None of the cultivars was resistant to any of the isolates used.Frequency of intermediate reactions among the cultivars ranged between 0.21-0.54. The cultivar with most intermediate reactions to the isolates used, OryzicaLlanos 5, is known for its high susceptibility under commercial fields. The resultssuggest a possible isolate-cultivar interaction with the existence of races thatdeserve more investigation in the future in order to facilitate the identification ofresistance genes.

The wild rice specie O. rufipogum exhibited a resistant or intermediate reaction tomost isolates tested (80%) while the species O. barthii and O. glaberrima weresusceptible to most isolates (table 2.8). Among the reported resistance sources

55

tested, the cultivar Remadja was resistant or intermediate to 80% of the isolatestested while the reported source Tetep was resistant to only 17% of the isolates.These potential sources will be tested in further studies in the greenhouse as wellas in the field. Interspecific populations between the species O. rufipogum and O.sativae have been developed through backcrossing and will be tested for theirreaction to sheath blight as well as used in the identification of molecular markersassociated with the resistance observed.

The results also showed that most of the advanced rice lines being tested inColombia or other Latin American countries for future release, as commercialvarieties are susceptible to sheath blight (table 2.9). These evaluations indicatethe need for the identification of suitable resistance sources to be incorporated inour breeding program, or the introgression of sheath blight resistance genespresent in wild species such as O. rufipogum. Greenhouse evaluations of USArice line/cultivars reported as resistant to sheath blight showed that thesecultivars are susceptible to the pathogen population present in Colombia (table2.10). These cultivars were either susceptible or highly susceptible to the isolatestested at two inoculation dates.

We are in the process of initiating a population genetic study of R.solani isolatesin Latin America using single –copy nuclear RFLP markers developed byRosewich et al. Dr. Bruce McDonald from ETH in Switzerland has agreed toshare with us those markers for our studies. Data from their studiesdemonstrated that the sheath blight pathogen in a Texas population is activelyoutbreeding (heterothallic) with high levels of gene flow and heterozygoteexcess. This means that the sheath blight fungus forms large amounts of novelgenotypes by sexual recombination contributing towards its continued successas a pathogen. Successful individuals increase by asexual reproduction to highfrequency. In order to develop a sound-breeding program for resistance to thispathogen, we will need to characterize the genetic structure and virulencediversity of this pathogen in the Latin American region.

2.A.5. Evaluation of Breeding Populations Incorporating ComplementaryResistance Sources to Blast in Greenhouse and Field Experiments

F.Correa and E.Tulande

We have initiated a study to evaluate the stability of blast resistance in advancedlines developed in a breeding program, where parents and crosses are selectedon the basis of their reaction to blast lineages and/or field reaction. A total of 309F2 blast resistant plants were selected in year 2000, out of 69 crosses and 169families (table 2.11). Resistant plants were selected out of triple crosses involvingthree resistant parents (greenhouse and field evaluations), three susceptibleparents, and crosses where the predominant F2 family field reaction was eithersusceptible, segregating (equal amount of F2 susceptible/resistant plants), orresistant F2 plants (table 2.11).

56

Selected F3 lines will be evaluated in year 2001 and 5 individual blast resistantplants will be selected within each one of the lines. Studies of the stability of theblast resistance reaction will be initiated in replicated trials in the F4 generationand continued in more advanced generations on a year basis until the lose ofresistance reaches a plateau. Our hypothesis is to demonstrate that those linesoriginating in crosses, where the F2 shows a higher number of blast resistantplants, and which showed a higher number of resistant sister lines, will give originto more stable resistant lines in the advanced generations. On the contrary,those advanced lines originating from F2 resistant plants selected within crosseswhere F2 susceptible plants predominated, will be less stable.

Our reasoning behind this hypothesis is that F2 populations exhibiting apredominant number of F2 resistant plants carry a larger number of differentresistance genes including minor genes. Advanced resistant lines originating inthese populations will probably carry all these resistance genes being morestable. On the contrary, the resistance observed in few F2 plants where most ofthe F2 plants are susceptible, would be controlled by just few resistance geneswhich would be easily defeated by the pathogen in early generations. Ourpurpose is to be able to recommend breeders to make detailed observations inthe F2 populations eliminating those crosses where F2 susceptible plantspredominate within the families, concentrating all efforts just in those crosseswith better chances to produce more stable blast resistant rice lines. Parentsused in the crosses under study are being planted for detailed observation oftheir blast reaction and for collection of blast isolates to be used in greenhouseinoculations.

2.A.6. Studies on the Interaction Effect of Macronutrients in theDevelopment of Rice Diseases

F.Correa, D.Delgado, E.Tulande, and P.Guzman

A three-year experiment to determine the effect and interaction of macronutrientson the development of rice blast, brown spot, leaf scald, sheath blight, and graindiscoloration was initiated in 1999, in collaboration with FEDEARROZ and FLAR.Doses of each nutrient considered in the different treatments are Nitrogen (0, 60,120,180 Kg/ha); Phosphorus (0, 40, 80, 120 Kg/ha); and Potassium (0, 40, 80,120 Kg/ha). The varieties used in the experiments include Oryzica 1, Fedearroz50 and Oryzica Yacu 9 depending on the site where the experiment is beingcarried out. Data for the years 1999 and 2000 has already been collected and isin the process of analysis. Fertilization levels used in year 2000 at Santa Rosafor favoring the development of the different rice diseases was based on theresults observed in 1999. Fertilization level recommended for the evaluation ofmost rice diseases at Santa Rosa was 180 KgN, 120 Kg P, and 80 Kg K.Although blast development was high in 2000 under this fertilization, otherdiseases such as brown spot or grain discoloration were low compared to year1999.

57

In another fertilization experiment aiming at identifying sources of resistance tograin discoloration, 104 rice lines selected in 1999 (74 resistant and 30susceptible) were evaluated under three different levels of fertilization todetermine the stability of their reaction at Santa Rosa (table 2.12). A total ofseven rice lines exhibited a resistant grain discoloration reaction of 1-3 in sixobservations under the three levels of fertilization (table 2.12). The most resistantline was CT 11891-2-2-7-M = Linea 30, which is an upland rice line beingreleased for the acid soils of the Colombian Altillanura. Other resistant lines(Caiapo, Progreso, Primavera, and IAC 47) are being developed for uplandconditions. Within this group of seven resistant lines, two of the irrigatedgermplasm developed by FLAR (Fl 00518-16p-8-2p and Fl 00440-47p-16-1p),also exhibited a resistant reaction to grain discoloration. In general, the japonicagermplasm is more adapted to the upland acid soils while indica germplasm ismore adapted to the irrigated conditions of Tropical Latin America. Ten otherlines exhibited a resistant to intermediate reaction (1-5) to grain discoloration(table 2.12). Four of these lines (Maravilla, Araguaia, Oryzica Sabana 10 andLinea 6) are upland materials developed for the Altillanura of Colombia or for theupland savanna soils of Brazil. Susceptible checks always exhibited a reactionscore of 9 for all observations in the three different fertilization treatments.

In general, our observations indicate that the upland germplasm exhibits highlevels of resistance to grain discoloration. This disease has increased inimportance in all rice growing areas of tropical Latin America in the last five yearsand farmers, are using high amounts of fungicides to control the disease. Effortsshould be made to introgress resistance of the upland germplasm into theirrigated rice cultivars.

2.A.7. Incorporation of Blast Resistance Genes into Commercial Varietiesthrough Backcrossing

F.Correa, G.Prado, and L.E.Berrio

Rice varieties are not adopted by farmers because of their high yield andexcellent grain quality and they remain in use when they are resistant to pestand diseases. In many instances, farmers would like to keep susceptiblevarieties due to the high yield and grain quality, preferring to use severalapplications of high amounts of fungicides.

We have initiated a backcrossing program in order to introduce blast resistanceinto some of those susceptible cultivars, which still play an important role in theeconomy of many rice farmers in Latin America. As an example, table 2.13shows a backcrossing program for the introgression of blast resistance into thesusceptible rice cultivars Epagri 108 and Epagri 109 from Brazil. Resistance isbeing derived from the resistant cultivars Fedearroz 50 and Oryzica Llanos 5.Blast isolates fully compatible with the susceptible cultivars were identified andisolated from these cultivars at Santa Rosa. They are being used in theevaluation under controlled conditions in the greenhouse for the selection andtransplanting of resistant plants for the production of BC2F1 plants used in future

58

inoculations. A large number of plants are being transplanted in order to selectthe most appropriate plants used in the backcross for the production of the BC2population. Other resistance genes, such as Pi-1, Pi-11 and Pi-2 are beingintrogressed through backcross into other susceptible commercial rice varieties.We expect to use this backcross technique on a more regular basis with the helpof molecular markers associated to the resistance genes Pi-1, Pi-11, and Pi-2.

2.A.8. Agropolis Advanced Research Platform CIAT-CIRADF.Correa and D.Tharreau

In order to understand the virulence/avirulence interactions of the rice blastpathogen and the rice plant for the development of suitable strategies forobtaining durable blast resistance, a three-year project within the AgropolisAdvanced Research Platform was initiated in 2000. The title of the project is:Understanding Deleterious Effects Caused by Avirulence/Virulence GeneCombinations to Developing Durable Resistance to the Rice Blast Disease.Analysis of genetics of virulence/avirulence gene combinations was undertakenusing fertile strains of the fungus existing in the CIRAD blast pathogen collection.Particular emphasis is placed on relevant gene combinations and geneticcrosses with Colombian isolates. Production of single isolates with virulence onthe resistance genes Pi-1, Pi-11, and Pi-2 is being pursued.

A total of sixty blast isolates from Asia, Africa, and Latin America were used inthe first part of this study. Many of the Asian isolates are highly fertile andsuitable for genetic crosses. Blast isolates exhibit compatibility with theresistance genes Pi-1 and Pi-11 or with Pi-2, but not with all three resistancegenes. Suitable crosses among isolates and inoculations on rice cultivarscarrying different resistance genes were performed under controlled conditions atCIRAD. A summary of the first progress report and conclusions based on theresults obtained this year is only presented in this chapter:

• The lack of compatibility of single isolates with the isolines C 101 LAC andC101 A 51 seems to be a common phenomenon, as shown by the analysis ofa blast population representing a wide geographic region of the world as wellas fertility in the blast pathogen.

• The above conclusion indicates that the cultivars C 101 LAC and C 101 A 51carry resistance genes that in combination should confer a more durable blastresistance.

• These studies indicate that the near isogenic line C 101 LAC carries, besidesthe resistance gene Pi-1, the resistance gene Pi-11, and the cultivar C 101 A51 carries the resistance gene Pi-2. Previous reports had indicated thepresence of only Pi-1 in C 101 LAC. Durability of blast resistance could thenbe developed by the combination of the resistance genes Pi-1, Pi-2, and Pi-11.

• It was possible to detect blast isolates compatible with each one of the genesPi-1, Pi-2, Pi-11, or combinations of any two of the genes. However, no single

59

isolate was virulent on the three genes. These results support our hypothesisthat the blast pathogen can not lose the three corresponding avirulencegenes as such occurrence might be deleterious for the pathogen.

• More crosses among blast isolates need to be developed in order to obtainisolates compatible with the three genes Pi-1, Pi-2, and Pi-11, to test theeffect of losing the corresponding avirulence genes in the pathogen and testthe potential durability of combining the three genes in a rice cultivar.

• Research should be initiated for transformation experiments of the blastisolates TH 16 (Thailand) and CL 73 (Colombia), compatible on theresistance genes Pi-2 and Pi-11, in order to disrupt the function of theavirulence gene Pi-11 giving origin to potential isolates compatible with thethree resistance genes. These isolates would allow us to study the effects oflosing the three avirulence genes on the pathogen. This approach is logic asCIRAD has already cloned and sequenced the avirulence gene for Pi-11.

• Research should be initiated in order to identify suitable molecular markersassociated with the resistance genes Pi-1, Pi-2, and Pi-11. These markerswoud be useful in the establishment of a marker assisted selection programfor the incorporation of the three genes in breeding lines aiming at having amore durable blast resistance. This approach can be initiated soon, sinceCIAT has already developed segregating populations of the cross between C101 LAC and C 101 A 51.

60

Table 2.1. Virulence Spectrum of Pyricularia grisea Isolates Collected inEvolution Studies of the Blast Pathogen

Genetic Lineage/Number of IsolatesResistance

GeneSRL 6B

3SRL 6

7SRL 4

23SRL 5

(n)2

1 Fukunishiki z, sh + + ++2 Fujisaka 5 i, kS ++ +++ +3 Aichi asahi A +++ +++ +++ ++4 Bl 1 b ++ ++ +++5 Bl 8 b +++ +++ +++ +6 Toride 1 zT ++ ++7 K 3 kh ++ +8 K 59 t + +++ ++ +9 K 60 kp + ++ ++10 F 80-1 k +++ +++ +++ +11 F 98-7 km +++ +++ +++ ++12 F 124-1 ta +++ +++ +++ +13 F 129-1 kp +++ +++ +++ ++14 F 128-1 Ta2 +++ + ++1 ++15 F 145-2 b ++ ++ +16 Zenith Z, a + + +17 Pi No.4 ta2 sh + +118 Rico ks +++ ++ +++ +19 Norin 22 sh ++ ++ ++20 Nato I + ++ +++ ++21 Shin 2 ks, sh ++ ++ +++22 Kanto 51 k +++ ++23 Tsuyuake Km ++ +++ ++24 Nipponbare sh +++ ++25 Ou 244 z ++ + +++26 Ishikari shiroke i, ks ++ ++ ++27 C 101 LAC 1,11 ++28 C 101 A51 2 +++ +++29 C 104 LAC 1 +++ ++30 C 101 PKT 4=ta ++ +++ +++ ++31 C103 TTP 1 +++ +++32 C104 PKT 3 +++ +++ +++33 C105 TTP1 4=-ta +++ +++ +++ ++34 C 105 TTP 4 (L23) Ta,+ +++ +++ +++ +++

Empty = 0% leaf area affected (LAA) +++ = Three isolates collected from Ceysvoni were compatible+ = 1-5% LAA 1= Three isolates collected from Ceysvoni were compatible++ = 6-30% LAA 2 = n=more than 10 isolates

61

Table 2.2. Cultivar of Origin Genetic Lineage and Number of Pyriculariagrisea Isolates Collected in Studies on the Evolution of the Blast Pathogen

Cultivar of Origin of Isolates

Genetic Lineage Oryzica 1OryzicaCaribe 8 Ceysvoni

OryzicaLlanos 5 IR 36 Total

SRL 6B 3 1 4

SRL 6 15 17 32

SRL 4 13 34 21 10 2 80

TOTAL 31 34 39 10 2 116

Table 2.3. Compatibility of Pyricularia grisea with the Cultivar of Origin inStudies on the Evolution of the Blast Pathogen

Origin of Isolates

Genetic lineage Oryzica 1OryzicaCaribe 8 Ceysvoni

OryzicaLlanos 5 Total

n. compatible isolates / n. isolates inoculated

SRL 6B 2/2 1/1 3/3

SRL 6 3/3 5/5 8/8

SRL 4 4/4 9/9 1/5 0/4 14/22

Total 9/9 9/9 7/11 0/4 25/33

62

Table 2.4. Identification of Blast Isolates Compatible with the Rice CultivarFedearroz 50 in Santa Rosa

Seed Blast Isolate (origin Fedearroz 50)No. Cultivar Source F 50-15-1 F 50-16-1 F 50-24-1

1 Fedearroz 50 Guamo HS S HS2 Fedearroz 50 Ibague S I HS3 Fedearroz 50 Piedras S I S4 Fedearroz 50 Monteria HS S HS5 Fedearroz 50 Valledupar S S HS6 Fedearroz 50 Saldaña I S S7 Fedearroz 50 CIAT basica 98 S S S8 Fedearroz 50 CIAT basica 99 S S I9 Fedearroz 50 Santa Rosa S S I10 Fedearroz 50 Patologia HS S S

11 Oryzica Llanos 4 CIAT HS - HS12 Oryzica Caribe 8 CIAT HS HS HS13 Oryzica Yacu 9 CIAT HS HS HS

14 C101LAC ( Pi-1,Pi-11) CIAT R R R

R= Resistant; I= Intermediate; S= Susceptible; HS= Highly Susceptible

63

Table 2.5. Rice Lines Exhibiting a Highly Resistant Leaf and Panicle BlastReaction (1-3) at Santa Rosa

Number Rice Line

1 CT 8455-1-24-3P-1X2 CT 8008-3-12-3P-1X3 CT 8008-16-10-3P-1X4 CT 8238-6-13-1P-1X5 CT 8250 –13-1-4P-1X6 CT 8250-18-4-4p-1x7 CT 8455-1-24-1P-1X8 CT 11014-10-1-29 CT 9737-1-1P-2-1

10 CT 11891-2-2-7-M11 VSTA/LBNT//RSMT12 IR 21015-72-3-3-3-113 GFMT *2/TQNG14 CT 10166-16-1-2P-1-315 CT 9509-17-3-1-1-M-1-3P-M-116 CT 11008-12-3-1M-4P-4P17 CT 10310-15-3-2P-4-318 IRGA 234-21-5-6-119 IRGA 660-3-13-5-320 CT 8455-1-13-1-M-2P21 CT 11280-2-F4 12P-522 CT 8945-8-17-2T-M23 CT 11256-5-F4-28P-5P24 CNAx5013-13-2-2-4-B25 CT 11369-1-F4-17P-4P26 CT 11626-14-4-2-1-M27 CNAx5013-12-13-2-2-4-B28 CT 10627-2-16-1T-1P-3P-429 CNARR 4955-7B-BM70A-45-5P

64

Table 2.6. Rice Lines Resistant to Lineages SRL-1 to SRL-6 of Pyriculariagrisea in the Greenhouse Exhibiting a Resistant Field Reaction

Blast Reaction (1-9)Rice Line Leaf Panicle

FL 00510-17P-3-2P-M 3 3FL 00447-27P-3-1P-M 2 3FL 00448-3º0-4-2P-M 2 3FL 00470-29P-2-3P-M 2 3FL 00470-29-P-5-2P-M 2 3FL 00470-29P-6-2P-M 2 1FL 00470-29P-7-3P-M 3 1FL 00585-12P-7-3P-M 3 1FL 00593-6P-1-3P-M 3 1FL 00593-6P-5-3P-M 3 1FL 00593-6P-7-1P-M 3 1FL 00593-6P-9-1P-M 3 1FL 00594-5P-9-3P-M 3 1FL 00594-5P-11-2P-M 3 1FL 00595-2P-1-1P-M 3 1FL 00595-25P-9-3P-M 3 3CT 13501-M-2-1-M-2-1P 3 3CT 13501-M-3-1-M-2-1P 3 1CT 13503-M-3-1-M-2-4P 2 1

Field reaction1-3 = Resistant; 4-9 = Susceptible

65

Table 2.7. Rhizoctonia solani Isolates Used in Greenhouse Evaluations ofSheath Blight

Isolate Identification Cultivar of Origin Site of Collection Year of Collection

1 Tolima 1953 –13 Oryzica 1 Saldaña 19882 Tolima 2061-1 Cica 4 Ambalema 19883 Tolima 2599 Fedearroz 50 Ibague 19984 Tolima 2064-1 Oryzica 1 Campoalegre 19885 Tolima 2313-6 Oryzica 3 Ibague 19906 SR 2156-1 Esparcidor Santa Rosa 19887 SR 2398 O.Llanos 4 Santa Rosa 19938 Tolima 2399 Oryzica 1 Tolima 19939 Tolima 2631 Oryzica 1 Purificación 199910 SR 2328 Unknown Santa Rosa 199011 Tolima 1953-2 Oryzica 1 Saldaña 198812 Tolima 2065-1 Oryzica 3 Campoalegre 198813 Tolima 2599 Fedearroz 50 Ibague 199814 Tolima 1954-3 Cica 4 Saldaña 1988

Rhizoctonia Isolate FrequencyCultivar 1 2 3 4 5 6 7 8 9 10 11 12 13 14 R-I

1 O.Llanos 5 S S S S S I S I - I I I I I 0.54

2 Oryzica 3 S S S S I S I I I I S S I I 0.50

3 Fedearroz 50 S S S S S S I I S I S S I I 0.36

4 Oryzica 1 S S I HS S S I I S S I I S S 0.36

5 Cica 8 I S S S S S S R I S S S I S 0.29

6 IR 22 I S S HS I S I HS S HS - S S S 0.23

7 O.Llanos 4 I S S HS S I S S R S S S S S 0.21

8 Selecta 3-20 S S S S S S S S R I S S S I 0.21

HS = Highly susceptible. Score 7.1–9. Lesions on the flag sheath or sheath #1S = Susceptible. Score 5.1-7. Lesions on sheath below leaf 2I = Intermediate. Score 3.1-5. Lesions on sheath below leaf 3R = Resistant. Score 1.1-3. Lesions on sheath below leaf 4HR = Highly resistant. Score 0-1. Lesions below leaf 4

66

Table 2.8. Greenhouse Evaluations of Wild Rice Species and ReportedSources of Resistance to Rhizoctonia solani

Cultivar Number Isolates Frequency R-I

Oryza rufipogon 10 0.80

Oryza glaberrima 9 0.11

Oryza barthii 9 0.0

Remadja 5 0.80

Pankai 6 0.50

Tadukan 6 0.33

Ta-Po-Choo-Z 6 0.17

Tetep 6 0.17

R-I – Resistant-Intermediate Reaction

67

Table 2.9. Greenhouse Evaluation of Advanced Rice Lines to Rhizoctoniasolani

Advanced Line Number of Isolates Frequency I – R

1 CT 9901-3-2-M-3-M-1 11 0.55

2 IR 51471-B-B-2-B-B-1-1 11 0.45

3 FB 100-10-1-M-1-M 11 0.36

4 CT 10192-5-1-2-CT-11 11 0.36

5 CT9509-17-7-1P-1PT 11 0.36

6 CT10240-10-1-2-1T-2-2 11 0.27

7 CT10310-15-3-2P-4-3 11 0.18

8 CT 10310-1-2-2T-1F-4P 10 0.10

I – R = Intermediate – Resistant Reaction

68

Table 2.10. Greenhouse Evaluation of USA Rice Lines with ReportedResistance to Sheath Blight

Sheath BlightLine Identification 30 DAP 65 DAP

1 NWBT/3/CBNT/9902/LBLE S HS

2 GFMT*2/TON6 HS HS

3 291643/MARS I S

4 20001*5/LMNT S HS

5 JEFFERSON HS HS

6 NWBT/KATY/LZOR/402003 S S

7 GCHW/RV83003116/LMNT/3/KATY S HS

8 KBNT HS HS

9 LMNT HS HS

10 CPRS S HS

11 RSMT S S

12 LBNT/STBN//NWBT/3//L202 HS S

DAP = Inoculation at 30 and 65 days after plantingI = Intermediate, S = Susceptible, HS = Highly susceptible.

69

Table 2.11. Selection of F2 Blast Resistant Plants in Crosses BetweenResistant or Susceptible Parents, and Selections Based on F2 Family FieldReaction

Crosses FamiliesResistant Plants

SelectedCross/Family (No) (No) (No)

Progenitors selected

R/R//R 13 27 53

s/s//s 8 15 28

F2 Family Field Reaction

Susceptible 27 50 89

Segregating 27 47 79

Resistant 18 30 60

TOTAL 69 169 309

R = Blast resistance, S = Blast susceptible

70

Table 2.12. Sources of Resistance to Grain Discoloration Evaluated UnderThree Fertilization levels at Santa Rosa in 2000

Fertilization level / Replication180 KgN 180 KgN 60 KgN120 KgP 0 KgP 80 KgP

GD 0 KgK 0KgK 80KgKNo. Rice Line 1999 I II I II I II

1 CT 11891-2-2-7-M R 1 1 1 1 1 32 Caiapo R 1 1 1 1 3 13 Progreso R 1 3 1 3 1 14 Primavera R 1 3 1 3 3 15 Fl 00518-16p-8-2p R 1 3 3 1 3 36 IAC 47 R 1 3 3 1 3 37 Fl 00440-47p-16-1p R 3 3 3 1 3 38 Maravilla R 1 5 1 3 1 19 Araguaia R 3 5 3 3 1 310 Fl 00440-47p-4-1p R 3 5 1 3 1 311 Fl 00440-47p-2-3p R 5 3 3 3 3 312 Fl 00558-26p-3-2p R 5 3 3 3 3 313 Fl 00440-47p-5-1p R 5 3 3 3 3 314 Fl 00448-9p-1-3p R 5 3 3 3 3 315 Oryzica Sabana 10 R 1 5 3 1 5 116 Linea 6 R 5 3 3 5 3 117 Fl 00478-29p-23-3p R 3 5 3 5 3 3

Susceptible controlCT 13432 S 9 9 9 9 9 9

1= Highly resistant; 9= Highly susceptible; R= resistant (Scale 1 – 3)GD= Grain Discoloration

71

Table 2.13. Backcrossing Program for the Introgression of BlastResistance to the Rice Cultivars EPAGRI 108 and EPAGRI 109

Resistant Plants TransplantedCROSS BACKCROSS BC1F1

FL 03020 E 108 // E108 / O. Llanos 5 26

FL 03021 E 108 // E 108 / Fedearroz 50 54

FL 03022 E 109 // E 109 / O. Llanos 5 17

FL 03023 E 109 // E 109 / Fedearroz 50 51

A total of 160 BC1F1 plants of each cross were inoculated with 8 isolates (20 plants each) compatible with Epagri 108 (E108) and Epagri 109 (E 109).

2.A.9. M.A.S. in RiceC. Quintero, F. Escobar, C. Martínez, F. Correa and J. Tohme

2.A.9.1. Background

Although a number of varieties resistant to both insect and virus are available,looking for molecular markers linked to the resistance genes would be useful inthe breeding program for a rapid development of new lines.

2.A.9.2. Materials and Methods

Among rice germplasm 159 lines were chosen because of their consistentreaction to the virus and its vector through several evaluations. DNA wasextracted individually and then eight groups (bulks) were conformed according toresistance and origin (table 3).

To evaluate a large set of microsatellites for M.A.S. in brachiaria.To continue with RAPD screening in rice and cloning polymorphic DNAfragments linked with the resistance to Sogata and HBV.

RAPD technique was used and approximately six hundred 10-meroligonucleotide primers were amplified in the eight DNA bulks. Until September2000, 330 primers were electrophoresed. For Sogata five RAPD bands wereconsistently found in all the resistant bulks (Figure 4) and 27 in the susceptibleones. When two of the primers showing resistant bands were screened in eachof the resistant varieties, only CT8447-5-6-3P-1X did not have the OPZ-19marker. For HBV, also some RAPD markers linked with resistance or

72

susceptibility were found but results were not as consistent as Sogata’s.This work is still in process.

Figure 2.1. Evolution Studies of the Rice Blast Pathogen

73

Figure 2.2. Amplification of OPZ-19 RAPD primer in rice varieties with differentlevels of resistance to Sogata dn HBV.

Publications

Referee:Seebold, K.W., Datnoff, L.E., Correa-Victoria, F.J., Kucharek, T.A., and Snyder,G.H. 2000. Effect of silicon rate and host resistance on blast, scald, and yield ofuoland rice. Plant Disease 84:871-876

Seebold, K.W., Kucharek, T.A., Datnoff, L.E., Correa-Victoria, F.J., andMarchetti, M.A. 2000. The influence of silicon on components of resistance toblast in susceptible, partially resistant, and resistant cultivars of rice.Phytopathology 00:000-000 (to be published in the December issue).

Peever, T.L., Zeigler, R.S., Dorrance A.E., Correa-Victoria, F.J., and Martin, S.2000. Pathogen population genetics and breeding for disease resistance.APSnet feature July 1-July 31, 2000.

Zeigler, R.S., and Correa-Victoria, F.J. 2000. Applying Magnaporthe griseapopulation analyses for durable rice blast resistance. In: Pathogen populationgenetics and breeding for disease resistance. APSnet feature July 1-July 31,2000.

OPZ-19

RESISTANT BULKSTO SOGATA

SUSCEPTIBLEBULKS TO

λλλλ PST

74

Non-refereed:

Prado, G.A., Correa-Victoria, FJ., Aricapa, G., Tulande, E., y Escobar, F. 2000.Hipotesis de la exclusion de linajes, una alternativa para el desarrollo decultivares de arroz con resistencia durable a Pyricularia grisea (Sacc) enColombia. Fitopatología Colombiana 23 (2): 54-58

75


2.B. Genetic Rice Resources Improvement for Latin America and theCaribbean. Sub Project - CIAT-CIRAD Collaborative Project

2.B.1. Characterization and Using Partial Resistance for the Control of RiceBlast

M.Vales, E.Tulande, J.Dossmann, M.C.Duque


CIAT-CIRAD collaborative project proposes a complete group of methods fordurable blast resistance selection. However, it should be tried, as a routine, toimprove these methods. Thus, a study of assisted selection with molecularmarkers and greenhouse selection for partial resistance was started.

Outputs

2.B.2. Study of partial blast resistance QTL

Study of partial blast resistance QTLS on lines IR 64/ Azucena was delayedbecause seeds multiplication was needed. Therefore the followingmethodological trials were made:

2.B.2.1. Field Study

It is already known that there can be varieties-strain interaction for partial blastresistance in the field (Rice Project Annual Report, 1999). A trial was carried outto observe possible varieties - nitrogen levels interactions for partial blastresistance in the field. To date we are analyzing the data.

2.B.2.2. Greenhouse Study

2.B.2.2.1. Introduction

To evaluate rice partial blast resistance in the field it is necessary to inoculatetrials with a compatible strain (Vales, 1989). Methods are mastered (Vales,1991; Vales et al., 2000b) but it is a heavy work. A good option to alleviate thisfieldwork would be greenhouse evaluation. Then the correlation between fieldand greenhouse for blast partial resistance was studied (Dossmann et al., 2000a,Dossmann et al., 2000b).

2.B.2.2.2. Materials and Methods

Two fungus strains of lineages SRL-6 and SRL-4 from Santa Rosa Experimental

76

Station, Meta, Colombia, were used. Greenhouse evaluations were made out ofinoculations in 23 rice Latin American varieties that did not show completeresistance to these strains. Partial resistance of these varieties in field conditionsto these strains is previously known (Rice Project Annual Report 1999).

Inoculations were done by manual aspersion to 18 days old plant, then startedthe incubation to materials in plastic dew chambers for 16 hours, after whichmaterials were withdrawn and placed under greenhouse conditions for 8 days.After this period plant symptoms are evaluated, taking into account mainly, typesof lesion and affected leaf area. To study correlation a statistic analysis wascarried with results previously taken in the field and with those found in thegreenhouse.

2.B.2.2.3. Results

Analysis led us to conclude that it does not exist a correlation between the fieldand greenhouse results. Thus, it is not exact to predict partial resistance of amaterial with respect to a strain in field, from data obtained in the greenhouse.

2.B.2.2.4. Perspectives

Evaluation of partial resistance in greenhouse does not represent resistance inthe field, that with varieties and trials in complete blocks. Then greenhouseevaluation can not help selection in segregant lines and with a statistic deviceless powerful adapted to the number of vegetable material, such as Federer’sblocks. Greenhouse evaluation does not have either interest for QTLS studies ofpartial blast resistance. Then we are obliged to evaluate partial resistance in thefield.

2.B.3. Use of Improvement Strategies for Durable Blast Resistance


Since the 1920s, selection to improve rice blast resistance is carried out with noreal success. Very little varieties have shown durable resistance. It was obtainedrandomly, more than as a result of improvement strategies, because it was notpossible to reproduce these results. Now it should be admitted that there is nomiraculous solution and that all improvement methods and all available geneticstrategies to combat this disease altogether should be used.

Different genetic strategies against rice blast disease are:

• Complete Resistance Utilization:

- Strategy 1: Use genes of other species.Complete resistance genes of other species could confer a durable

77

protection. Several research teams are working in rice transformation withsuch genes: CIAT, Dra. Z. Lentini; CIRAD with the European project EURICE;etc.

- Strategy 2: Virulences Incompatibility.Some associations of few virulence genes, in a same strain, are of very lowfrequency in the world. Therefore, associations of corresponding resistancegene should confer rice a durable protection. It is not necessary, in that case,to know the lineages.

- Strategy 3: Accumulation of complete resistance genes using lineagesknowledge.

- Strategy 4: Selection of a general, poligenic, and partial resistance.

At a practical level to develop varieties:• Strategy 1: Use genes of other species.We are waiting for the first results of involved teams with transgenic. Then thereare no new available resistance progenitors.

• Strategy 2: Virulences incompatibility.In the International Platform of Montpellier, France, Dr. Fernando Correa, inCIRAD’s Rice Pathology Laboratory, is studying the existence of virulencesincompatibility. Meanwhile recurrent populations are made with involved genes ofcomplete resistance (Cf. 2,3. Constitution of new populations […] / 2.3.2.Populations source of complete resistance genes).

• Strategy 3, accumulation of complete resistance genes using lineagesknowledge, and

• Strategy 4, selection of a general, poligenic and partial resistance.Since these two strategies are not miraculous, the idea is to use both and toaccumulate complete and partial resistance genes in the same plants. That iswhat we are going to consider below.

Outputs

2.B.4. Traditional Improvement Using Selected ParentsL.E.Berrio, D.Delgado, M.Vales

Rice blast is also a very serious potential problem for upland rice for hillsides.Because it is new crop, this disease is not present yet in Cauca’s hillsides.Therefore, the idea is to obtain some information about the resistance of 203upland rice for hillsides with their planting in Los Llanos, Santa RosaExperimental Station, Meta, Colombia. A limitation is that the pathogenic funguspopulation is specific to indica rice and not to japonica of hillsides.

78

On the other hand, these lines can be of FLAR’s interest that is seeking formaterial with cold resistance for some Colombia’s zones and for the LatinAmerica South Cone.

The table 2.14 shows that lines were immune of leaf and neck Blast disease.Only the immunity may be considered, because differences of symptom levelcannot be significant in this trial condition (Rice Project Annual Report 1999).These lines come from a selection in cold altitude, so many are too tall. For thesame reason many are flowering very early. The minimum of sawing-floweringcycle is 56 days.Lines 3205 and 3206 have a long and thin grain.Although the population of Pyricularia of Santa Rosa is specifically adapted toindica, 17 japonica lines F5 have been selected for the breeding program ofupland rice of altitude.The table 2.15 shows that all these lines have a direct parent endowed of a highlevel of partial resistance.

2.B.5. Recurrent Selection to Improve Partial and Complete Resistance andOther Agronomic Traits in Population PCT-6

M.Vales, E.Tulande, J.Dossmann, M.Triana, V.Kury, M.C.Duque


Recurrent selection is the adequate method to improve poligenic traits. Thereforethis method was proposed for the first time in rice mainly for accumulation ofblast resistant genes (Vales, 1983 and 1987). To facilitate recombinations a malesterility recessive gene is used ( Singh and Ikehashi, 1991).

Developed Scheme for recurrent selection has three parts (Vales, 1998; Vales etal., 1999b):• Complete resistance selection.• Selection for partial resistance and other agronomic traits.• Genetic recombination that also allow to discard a too quick lost of population

variability and to maintain high frequency of the male sterile gene withparticipation of a not selected population sample (S0).

2.B.5.2. Materials and Methods

Indica irrigated population PCT-6 was selected twice for Rice Hoja Blanca virusbefore recurrent selection. It was a work of M.-H. Chatel, M. Triana and J.Borrero. After that recurrent selection was made (Rice Project Annual Report1999):

• Selection for complete resistance:- 1997B and 1998B: Two selection cycles in greenhouse in S1 lines. Genetic

progress of complete resistance due to first selection cycle was great (Rice

79

Project Annual Report 1999, Vales et al., 2000a).- 1999A: A reciprocal selection cycle plant-parasite in field in S1 plus S3.

• Selection for partial resistance and other agronomic traits:- 1998B: Selection for partial blast resistance, yield, earliness, type of plant,

insects resistance (Diatraea saccharalis) in field in lines S2, in laboratory forS3 seeds grain quality and in greenhouse for resistance to Sogata(Tagosodes orizicolus) in S3 plants.

• Genetic recombinations:- 1998A: Recombination in S1 plus S0, after selection for complete resistance

in S1 in greenhouse.- 1999A: Recombination in S1 plus S3, and incorporation of S0, after selection

for complete resistance completes in S1 in greenhouse and selection forpartial resistance in S2 in field and other agronomic traits in S2 and S3.

• To obtain material for actual trial:- 1999B: S0 from last recombination was planted. Information on S0 plants with

"S1" or "S3" origin is kept. Within 5000 plants, as half SIB lines, a selection forhigh inheritance characters was made. Three hundred male fertile plantswere selected to obtain seeds from 300 S1.

A second recurrent selection cycle for both resistances and agronomic traitsbegin with:Complete resistance selection in greenhouse. This also gives the opportunity toevaluate genetic progress for complete resistance.Plants without complete resistance to a strain was selected. Progenies of theseplants are going to be used for partial resistance selection. Inoculation of 20plants, each of 300 lines S1, was made.

2.B.5.3. Results and Discussion

Genetic progress evaluation for complete resistance:Practically there is no difference for the complete resistance spectrum, in termsof percentage of incompatible strains, between S1 lines of origin "S1" or " S3".That demonstrates that genetic recombination through the use of a male sterilegene is very good.It can be observed that with a recurrent selection cycle, also for partial resistanceand other agronomic traits, the genetic progress for complete resistance is verystrong (Fig.2.3). There are already lines S1 that have all their plants withcomplete resistance to all inoculated strains of 7 lineage observed in theColombian Llanos.

2.B.5.4. Perspectives

These results allow adaptation of recurrent selection plan:

80

Practically, there is no difference, for complete resistance, among lines S1 "S1" or"S3". Then the following cycle of recurrent selection will be accomplished solelyfrom S1of origin "S3", to try to maintain more firmly genetic progress due toselection for partial resistance and other agronomic traits.Lines S1 that have all their plants, or plants, with complete resistance to allinoculated strains of 7 lineages will be used to create varieties and form the firstrecurrent population of correct projection.To create varieties it is necessary to use a particular selection scheme forcomplete and partial resistance (Vales et al., 1999a). Selection in the projectedpopulation will be drastic as it is in narrow genetic base population. It is apopulation that it going to be delivered to associates.

2.B.6. New Populations Constitution for Recurrent Selection of Partial andComplete Resistance, and Other Agronomic Traits

2.B.6.1. Recurrent Populations of Narrow Genetic Base

New concept of recurrent population of narrow genetic base (Vales et al., 1998)is used. In 2000A it was made the first genetic recombination in 5 populations ofthis type, made for several rice crop conditions (Rice Project Annual report1999). These populations were delivered to some of the country partnersaccording to their rice crop conditions:

Argentina University of La PlataBrazil CIRADChile National Agronomic Research Institute (INIA)China Agronomic Sciences Academy of Yunnan (YAAS)Colombia Federation of Rice Producers (FEDEARROZ)Ivory Coast West Africa Rice Development Association (WARDA)Costa Rica University of Costa Rica (UCR)El Salvador National Center of Agricultural and Forest Technology (CENTA)France CIRAD/ French Rice Center (CFR)Madagascar CIRAD/ National Center of Rural Development (FOFIFA).

2.B.6.2. Populations Source of Complete Resistance Genes

Analysis of any recurrent population could show that it lacks complete resistanceto an inoculated strain in the greenhouse. Reciprocal recurrent selection plant-parasite in field could arrive to a same conclusion. Therefore for not loosing theobtained genetic progress in such population, it is necessary to seek for the mostadequate method to introduce in it the resistance it lacks. That is why apopulation CRS is made, in English acronym for complete resistances source.This population has the male sterility gene and nearly all identified genes ofcomplete resistance.To take advantage of Strategy 2, virulences incompatibility, in recurrent selection,populations CRS that have the male sterility and genes of corresponding

81

complete resistance from the different models, are being formed.

Use of these CRS population is done through the adjunct of one(s) of thesample(s) into already available recurrent populations.

In the year 2000B F1 were planted. These CRS populations will be available forthe country partners by the end of 2001A.

2.B.7. Training and Publication of Results

Training is a very important part of a basically methodological research. Theactivities of this year were:• Animation of Seminar "Recurrent Selection for durable blast resistance and

other agronomic traits", Monday April 24,2000 in University of Costa Rica (UCR), Costa Rica. (Cf. Trip report).

• Participation in workshop Los Soles of CIAT’s Hillsides Program, April 25 to27, 2000, in San Dionisio, Nicaragua (Cf. Trip report).

• Animation of seminar on "Recurrent Selection for durable blast resistance andother agronomic traits", July 13, 2000, CIRAD, Montpellier, France.

• Participation in the Third International Congress 2000 CROP SCIENCE,August 17 to 22, 2000, Hamburgo, Germany. (Vales et al, 2000a; Cf thesummary in annex).

• Animation of seminar "Recurrent Selection for durable blast and otheragronomic traits", September 1st, 2000, at the Food and Crop ResearchInstitute/ YAAS, Kunming, Yunnan, P. R. China (Cf. Trip report).

• Participation in "Advanced course on integrated application of molecular andconventional methods in rice crop improvement", September 25 to October6,2000, at CIAT, Palmira, Valle del Cauca, Colombia, with a course on"Varietal Improvement seeking for desirable blast resistance" (Vales, 2000a;Cf summary in annex).

• Participation in International Symposium, Durable Disease Resistance: a keyto sustainable agriculture. Wageningen, The Netherlands, November 28-December 1, 2000 (Vales, 2000b; Vales et al, 2000b; Cf summaries inannex).

• Preparation of national course of rice selection in June 2001 in Cuba.• Training preparation for Peng Xi of Yunnan Academy for Agriculture

Sciences, P. R. of China (Cf. Results 1. Reinforcement of genes sources / B.Confronting food insecurity in the hillsides/ 2.2. New associates search).

• Joanna Paola Dossmann presented teams results during the followingevents:

- II Regional Agroscience and Technology Seminar, Century XXI, ColombianOrinoquia. CORPOICA, PRONATTA. August 23 – 25, 2000. Hotel del Llano,Villavicencio, Meta, Colombia. (Dosman et al, 2000a).

- XXI Congress of Colombian Association for Phytopathology and SimilarSciences (ASCOLFI): Pathology of post harvest in flowers, fruit bearing,vegetables, seeds, roots and tubers. Palmira, International Center of Tropical

82

Agriculture (CIAT), August 30 to September 1st, 2000. (Dosman et al,2000b).

2.B.8. Publication of a Manual on Selection of Blast Resistance

The last phase of this publication is advancing. This manual is expected to bepublished in 2001.

References• Dosman J.P., Vales M., Tulande E., Duque M.C. 2000a. Estudio de la

correlación entre la resistencia parcial al añublo (Magnaporthe grisea) encampo e invernadero. II Seminario Regional Agrociencia y Tecnología, SigloXXI, Orinoquia Colombiana. CORPOICA, PRONATTA. Agosto 23-25 del2000. Hotel del Llano, Villavicencio, Meta, Colombia. (Cf. Annexes).

• Dosman J.P., Vales M., Tulande E., Duque M.C. 2000b. Estudio de lacorrelación entre la resistencia parcial al añublo (Magnaporthe grisea) encampo e invernadero. XXI Congreso Asociación Colombiana de Fitopatologíay Ciencias Afines (ACOLFI): Patología de la postcosecha en flores, frutales,hortalizas, semillas, raíces y tubérculos. Palmira, Centro Internacional deAgricultura Tropical (CIAT), Agosto 30 a Septiembre 1 del 2000. (Cf.Annexes).

• Singh R.J., Ikehashi, H. 1991. Monogenic male-sterility in rice : introduction,identification and inheritance. Crop Science 21 : 286-289.

• Vales, M. 1983. From the knowledge of relationship plant-parasite to thestrategies against rice blast disease. PhD Thesis. Plant Amelioration andDevelopment. Univ. of PARIS SUD, Center of Orsay. May 2, 1983 (Fr):310 p.

• Vales, M. 1987. Durable resistance : case of rice blast disease. II – Breeding fordurable resistance. L'Agronomie Tropicale, 42 (2) (Fr): 112-120.

• Vales, M. 1989. Breeding strategy for rice blast resistance. InternationalSymposium on the Biology role for solution to the food crisis in Africa. AfricanBiosciences Network. PNUD-FAO. July 26-30, 1989, Yamoussoukro, IvoryCoast (Fr): 12 p.

• Vales, M. 1991. New breeding method for upland rice varieties with durableblast disease resistance. National Plant Protection Association (ANPP)- IIIInternational Conference on Plant Diseases, Bordeaux - December 3-5, 1991(Fr): 785-792.

• Vales, M. 1998. Recurrent selection for rice blast resistance. Differentschemes used in the collaborative project CIAT-CIRAD in III Upland RiceInternational Breeders Workshop, Goiania, Go. Brazil – March 1998: 4 p.

• Vales, M., Chatel , M.-H., Borrero, J., and Ospina, Y. 1998. Recurrent Selectionfor rice (Oryza sativa) blast (Magnaporthe grisea) Resistance in Populationwith Narrow Genetic Base. International Symposium on Rice GermplasmEvaluation and Enhancement, Aug. 30 – Sept. 2, Suttgart, Arkansas, U.S.A.

• Vales, M. J., Borrero, J., Ospina, Y., Tulande E., Gibbons, J., and Gonzales,D. 1999a. Selection strategy for durable rice blast resistance. Pedigreeselection for rice blast (Magnaprothe grisea) complete, and partial

83

resistances. The 2nd Template Rice Conference. June 13-17, 1999.Sacramento, California.

• Vales, M. J., Chatel, M.-H., Borrero, J., Delgado, D., Ospina, Y., Tulande E.,Triana, M., Meneses, R., Kuri, V., Duque, M. C., and Silva, J. 1999b.Selection strategy for durable rice blast resistance. Recurrent selection forrice blast (Magnaporthe grisea) complete, and partial resistances. The 2ndTemplate Rice Conference. June 13-17, 1999. Sacramento, California.

• Vales, M. 2000a. Breeding strategies for durable resistance to rice (Oryzasativa) blast (Magnaporthe grisea) disease. International Symposium,Durable Disease Resistance: a key to sustainable agriculture. Wageningen,The Netherlands, November 28-December 1, 2000 (Cf. Anexos).

• Vales, M. 2000b. Mejoramiento varietal buscando resistencia deseable a laPiricularia. In: Curso avanzado sobre aplicación integrada de métodosconvencionales y moleculares en el mejoramiento del cultivo del arroz. 25septiembre – 6 octubre, 2000, CIAT, Palmira, Valle del Cauca, Colombia: 60p.

• Vales, M, M.-H. Chatel, J. Borrero, E. Tulande, M. Triana, V. Kury, M.-C.Duque, Y. Ospina. 2000a. Breeding strategy for durable rice (Oryza sativa)blast (Magnaporthe grisea) resistance: recurrent selection for complete, andpartial resistance, and for other agronomic traits. In: 3rd international CROPSCIENCE Congress 2000. 17-22 August 2000, CCH, Hamburg, Germany (Cf.Anexos).

• Vales, M., Chatel M. H., Borrero J., Dossmann J., Tulande E., Triana M., KuryV., Duque M. C., and Ospina Y. 2000b. Durable resistance to rice (Oryzasativa) blast (Magnaporthe grisea) disease: Recurrent selection breedingprogram of the CIRAD/CIAT collaborative project. International Symposium,Durable Disease Resistance: a key to sustainable agriculture. Wageningen,The Netherlands, November 28-December 1, 2000 (Cf. Anexos).

84

Figure 2.3. Genetic Progress for Complete Resistance in Population PCT-6\HB Obtained with a Recurrent Selection Cycle.

30

25

20

15

10

5

030 40 50 60 70 80 90 10020100

Complete Resistance spectrum in percentages of incompatible strains

Perc

enta

ge o

f S1

lines

3 S1

Before the selectionAfter a complete selection cycle and recombination

85

Table 2.14. FLAR Evaluation Of 203 F5 Lines Of Upland Rice For HillsidesIn Santa Rosa Experimental Station, Meta, Colombia, 2000 A(Cf. Convention at the End)

NCAM PEDIGREE (SeeTab.2)

VG HT BL1 BL2 FL LSC BS NBL GD

3196 PRA505 3 113 1 1 78 3 5 3 73197 PRA506 3 72 3 2 74 1 3 1 33198 PRA506 5 125 2 2 59 1 1 3 13199 PRA506 5 123 3 3 62 1 1 3 13200 PRA510 5 124 4 4 77 1 3 1 53201 PRA515 3 127 4 5 78 3 3 5 53202 PRA515 3 120 4 4 77 3 5 3 53203 PRA515 3 114 4 4 77 3 5 1 53204 PRA519 3 134 4 4 68 3 3 1 3

CICA 8 1 4 7 --- - - - -ORYZICA 1 1 88 5 5 97 3 5 5 3CEYSVONI 5 79 4 4 85 1 3 1 1

3205 PRA522 5 112 1 2 73 3 3 1 53206 PRA522 7 115 2 2 73 1 3 1 33207 PRA522 5 123 4 4 71 1 3 3 53208 PRA522 3 120 4 3 82 3 3 3 73209 PRA522 3 113 4 4 82 3 5 1 73210 PRA522 7 100 3 3 78 1 3 5 33211 PRA522 7 123 4 3 78 1 3 5 33212 PRA522 5 102 3 2 74 1 3 5 13213 PRA522 5 115 3 3 77 1 3 5 33214 PRA522 5 121 3 2 77 1 3 1 33215 PRA522 5 123 3 3 77 1 3 3 53216 PRA522 7 91 3 3 77 1 3 7 33217 PRA522 5 119 3 4 78 1 3 3 53218 PRA522 5 100 4 4 71 1 3 3 33219 PRA522 5 115 4 3 71 1 1 3 13220 PRA522 5 92 3 4 74 1 3 5 33221 PRA522 7 87 6 5 78 1 3 3 33222 PRA522 - --- --- --- -- - - - -3223 PRA522 5 91 3 3 74 1 3 5 33224 PRA522 5 124 3 4 82 3 3 5 33225 PRA522 7 101 3 4 77 1 3 3 33226 PRA522 5 112 2 3 77 1 3 1 33227 PRA522 5 91 5 4 78 1 3 3 33228 PRA522 5 101 4 4 82 1 3 3 33229 PRA522 7 126 3 4 77 1 3 1 33230 PRA522 7 96 5 4 84 1 3 1 33231 PRA522 7 102 3 4 82 1 3 1 33232 PRA522 7 106 3 3 78 1 3 3 53233 PRA522 7 109 3 3 78 1 3 1 33234 PRA522 5 120 2 2 74 1 3 1 3

86

CICA 8 1 -- 4 7 --- - - - -ORYZICA 1 3 -- 5 5 98 3 5 5 3CEYSVONI 3 84 3 4 85 1 3 1 1

3235 PRA522 5 125 2 2 74 1 3 1 33236 PRA522 5 121 2 2 74 1 1 1 33237 PRA532 5 118 4 4 68 3 1 1 13238 PRA542 5 113 4 3 77 1 3 1 33239 PRA542 7 98 2 2 77 3 3 1 33240 PRA542 5 119 3 3 71 1 5 1 33241 PRA542 5 125 4 4 70 1 3 1 33242 PRA542 5 115 3 4 68 1 3 3 33243 PRA542 5 116 3 4 71 1 3 3 33244 PRA542 5 112 4 2 74 1 3 1 33245 PRA542 7 102 4 3 65 1 3 1 53246 PRA544 9 91 4 4 62 1 3 3 33247 PRA544 9 98 4 4 62 1 3 3 33248 PRA544 7 113 4 4 67 1 3 1 33249 PRA544 5 102 4 5 78 1 3 3 53250 PRA544 5 105 4 5 68 1 3 3 53251 PRA544 5 103 4 5 82 1 3 1 53252 PRA544 5 105 5 4 82 1 3 3 53253 PRA544 5 104 4 4 74 1 3 1 53254 PRA544 7 114 5 5 74 1 3 1 53255 PRA544 7 97 3 3 84 1 3 1 53256 PRA544 5 87 5 5 84 1 3 1 73257 PRA546 5 118 4 5 73 1 3 1 33258 PRA553 5 111 3 3 68 1 1 1 13259 PRA553 5 102 4 4 68 1 1 1 33260 PRA553 7 87 3 3 59 1 1 1 33261 PRA553 5 103 4 3 58 1 1 1 13262 PRA553 5 95 4 3 58 1 1 1 33263 PRA553 5 102 4 5 78 1 3 1 33264 PRA553 7 95 4 3 68 1 1 1 5

CICA 8 1 5 7 --- - - - -ORYZICA 1 3 5 5 98 3 5 5 3CEYSVONI 5 84 3 4 84 1 3 1 1

3265 PRA553 7 75 3 3 56 1 1 1 33266 PRA553 9 64 3 3 56 1 3 3 33267 PRA553 - --- --- -- - - - -3268 PRA553 - --- --- -- - - - -3269 PRA553 5 98 3 3 68 1 1 1 33270 PRA553 5 107 2 2 64 1 1 1 33271 PRA553 5 99 3 2 66 1 1 1 13272 PRA553 5 92 2 2 69 1 3 1 13273 PRA553 5 84 3 2 67 1 3 1 13274 PRA553 5 81 1 2 70 1 3 1 13275 PRA553 - --- --- -- - - - -3276 PRA553 5 86 2 2 68 1 3 1 33277 PRA553 - --- --- -- - - - -

87

3278 PRA553 - --- --- -- - - - -3279 PRA553 5 112 2 1 70 1 1 1 13280 PRA553 5 97 3 2 68 1 3 1 33281 PRA553 7 63 2 2 58 1 3 1 33282 PRA553 5 94 2 4 64 1 3 1 33283 PRA553 7 116 3 3 78 1 1 1 53284 PRA553 5 110 5 5 73 1 1 1 53285 PRA553 7 92 4 4 56 1 1 3 33286 PRA553 5 111 4 5 56 1 1 1 33287 PRA553 7 101 4 4 58 1 1 1 13288 PRA553 7 99 3 4 60 1 1 1 13289 PRA553 5 100 2 2 60 1 3 1 33290 PRA553 5 112 1 1 60 1 3 1 33291 PRA553 5 114 3 4 68 1 3 1 33292 PRA553 5 92 3 3 62 1 3 1 33293 PRA553 5 97 3 3 66 1 3 1 13294 PRA553 5 96 2 2 64 1 3 1 1

CICA 8 3 6 8 --- - - - -ORYZICA 1 1 84 5 5 96 3 5 3 3CEYSVONI 5 82 3 4 86 1 3 1 1

3295 PRA553 5 100 1 1 56 1 3 1 13296 PRA553 5 89 1 2 58 1 3 1 33297 PRA553 5 91 1 1 66 1 3 1 13298 PRA553 5 92 1 1 56 1 3 1 13299 PRA553 5 88 1 1 56 1 3 1 13300 PRA553 5 103 4 4 58 1 1 3 33301 PRA553 5 92 4 4 58 1 3 3 33302 HT JUMLI MARSHI 9 105 6 6 86 1 1 3 53303 PRA557 7 115 3 2 71 1 3 1 33304 PRA559 5 116 3 2 68 1 3 1 33305 PRA565 5 106 3 2 67 1 3 1 33306 PRA565 5 128 4 4 66 1 3 1 33307 PRA565 5 93 4 4 74 1 3 1 33308 PRA565 5 90 2 3 67 1 5 3 33309 PRA565 5 113 3 4 79 1 3 3 33310 PRA565 5 93 5 3 68 1 5 1 13311 PRA565 7 106 4 4 78 1 5 1 33312 PRA565 5 91 3 3 62 1 5 1 13313 PRA565 5 98 3 2 68 1 3 1 33314 PRA565 5 107 4 3 76 1 3 1 33315 PRA565 5 100 3 2 74 1 3 1 33316 PRA565 5 86 2 2 60 1 3 3 33317 PRA565 5 107 2 2 68 1 3 1 13318 PRA565 5 117 2 1 74 1 3 1 33319 PRA565 5 128 1 1 75 1 3 1 33320 PRA565 5 112 1 1 58 1 3 3 33321 PRA565 5 114 1 1 62 1 3 3 3

CICA 8 3 4 6 --- - - - -ORYZICA 1 1 4 5 98 3 5 5 3

88

CEYSVONI 5 81 3 4 86 1 3 1 1


3322 PRA565 5 96 1 1 58 1 3 3 33323 PRA565 5 87 1 1 68 1 1 1 33324 PRA565 5 91 1 1 64 1 1 1 33325 PRA565 7 99 4 3 60 1 3 3 33326 PRA565 7 106 2 2 68 1 3 1 33327 PRA565 7 108 2 2 68 1 1 1 33328 PRA565 7 115 2 2 70 1 3 1 53329 PRA565 7 107 2 2 75 1 3 3 33330 PRA565 7 102 2 2 71 1 1 1 53331 PRA565 5 118 1 1 62 1 3 1 33332 PRA565 5 104 1 1 62 1 3 3 33333 PRA565 5 97 1 1 62 1 3 3 33334 PRA565 5 110 1 1 60 1 3 3 33335 PRA565 3 123 1 1 67 3 3 1 13336 PRA565 5 86 3 3 70 3 3 3 33337 PRA565 5 116 3 3 71 3 3 1 33338 PRA565 5 96 3 3 71 3 3 1 33339 PRA565 5 101 3 4 71 3 5 1 33340 PRA565 5 93 1 4 67 3 3 1 53341 PRA565 5 99 3 4 71 3 3 3 33342 PRA565 3 94 3 4 69 3 3 7 13343 PRA565 3 108 4 4 72 3 3 5 33344 PRA565 5 97 4 3 71 1 1 1 13345 PRA565 5 109 3 3 70 1 3 1 13346 PRA565 5 107 3 4 71 1 1 7 33347 PRA565 7 98 3 4 73 1 3 5 33348 PRA565 7 97 3 4 73 1 3 3 3

CICA 8 1 5 7 108 3 1 7 3ORYZICA 1 3 5 5 98 3 5 5 3CEYSVONI 3 86 3 4 85 1 3 1 1

3349 PRA565 - 93 9 9 82 1 3 1 33350 PRA577 5 99 5 4 75 3 3 3 33351 PRA577 5 104 3 4 78 3 3 3 53352 PRA577 3 100 1 2 84 5 3 3 53353 PRA577 5 103 2 2 5 1 1 53354 PRA577 5 97 2 4 73 3 3 5 53355 PRA577 3 100 2 2 3 3 1 73356 PRA599 5 138 2 2 82 1 1 1 33357 PRA533 5 103 4 4 3 1 1 73358 PRA533 3 86 3 4 75 3 3 5 33359 PRA556 - 104 9 9 90 1 1 5 13360 PRA556 - --- --- -- - - - -3361 PRA560 3 97 2 2 85 3 5 5 33362 PRA560 5 83 4 4 91 1 3 1 3

89

3363 PRA560 5 94 4 3 91 1 5 1 33364 PRA560 5 91 3 3 88 3 3 3 33365 PRA560 5 78 3 3 87 3 3 1 33366 PRA560 5 93 3 2 85 3 3 1 33367 PRA560 5 85 4 4 88 3 3 1 53368 PRA564 5 135 4 5 84 3 3 3 33369 PRA8 5 117 4 4 67 1 1 1 13370 PRA523 3 115 4 4 70 1 1 1 33371 PRA523 5 105 4 4 71 1 1 1 33372 PRA523 5 120 3 4 70 1 1 1 33373 PRA523 5 133 2 4 70 1 1 1 33374 PRA524 5 95 3 4 64 1 1 3 13375 PRA524 5 112 3 4 73 1 3 3 53376 PRA524 5 130 4 3 68 1 3 1 33377 PRA524 5 107 4 4 67 1 3 1 33378 PRA524 5 131 4 4 62 1 3 1 3


3379 PRA524 5 120 3 3 67 1 3 1 33380 PRA524 5 120 3 3 67 1 3 1 33381 PRA620 5 125 4 4 66 1 3 1 33382 PRA620 5 126 4 4 66 1 3 1 33383 PRA620 5 114 4 4 67 1 3 1 33384 PRA620 5 124 4 4 66 1 3 1 33385 PRA620 3 130 4 3 64 1 3 1 33386 PRA524 5 148 4 3 78 3 3 1 13387 PRA524 5 143 4 4 78 1 3 1 33388 PRA524 9 105 3 3 60 1 3 3 33389 PRA622 5 111 3 4 66 1 3 1 33390 PRA622 5 118 4 3 64 1 3 1 33391 PRA622 5 122 3 3 72 1 3 1 53392 PRA632 5 109 4 4 77 1 3 3 73393 PRA632 5 107 4 4 77 1 3 3 53394 PRA632 5 96 3 3 74 1 3 3 53395 PRA632 5 101 4 4 77 1 3 3 73396 PRA632 5 105 4 3 78 1 3 1 53397 PRA632 5 103 4 4 77 3 3 3 53398 PRA632 5 100 4 4 74 1 3 3 73399 PRA632 5 100 4 3 73 1 3 3 7

Conventions (IRRI Scales; 1:better, 2:worst):VG: VigorHT: Height (cm)BL: Leaf Blast FL: flowering (days from sawing)LSC: Sheath Blight BS: Narrow Brown SpotNBL: Neck Blast GD: Grain discolorationIn gray: the selected lines for upland rice for hillsides program.

90

Table 2.15. Pedigree Of The F5 Lines Evaluated By The FLAR

Cross PedigreePRA8 Latsidahy / IRAT 351 = Latsidahy // Daniela / IAC 25PRA505 CA 148 / FOFIFA 133 = CA 148 // RS 25 T / DanielaPRA506 CA 148 / IREM 239PRA510 Chhomrong Dhan / IRAT 351 = Chhomrong Dhan // Daniela / IAC 25PRA515 Daniela / IAC 25 // IRAT 351 / Rikuto Norin 15PRA519 Dourado Precoce / FOFIFA 60 = Dourado Precoce // Daniela / IAC 25PRA522 PRA8 / IRAT 265-57-2 = Latsidahy / IRAT 351 /// IRAT 13 //Dourado Precoce / IRAT 13PRA523 Latsidahy/ IRAT 351 // Latsidahy = Latsidahy // Daniela / IAC 25 /// LatsidahyPRA524 FOFIFA 60 / Chhomrong Dhan = Daniela / IAC 25 // Chhomrong DhanPRA532 IRAT 352 / IRAT 265-57-2 = Daniela / IAC 25 // IRAT 265-57-2PRA533 FOFIFA 116 / AlicomboPRA542 IAC 25 / IRAT 380 = IAC 25 // Latsidahy / IRAT 351PRA544 IAC 25 / Daniela // IRAT 353 / Shin Ei = IAC 25 / Daniela /// Daniela / IAC 25 // Shin EiPRA546 IRAT 114 / IRAT 380 = Mut. Moroberekan // Latsidahy / IRAT 351PRA553 IRAT 265-57-2 / Jumli Marshi = IRAT 13 /// Dourado Precoce // IRAT 13 / Jumli MarshiPRA556 IRAT 265-57-2 / Latsidahy = IRAT 13 /// Dourado Precoce // IRAT 13 / LatsidahyPRA557 IREM 239 // Daniela / IAC 25PRA559 IREM 239 / KhonoralloPRA560 IREM 238 / KhonoralloPRA564 Khonorallo / AlicomboPRA565 Khonorallo / IRAT 265-57-2 = Khonorallo / IRAT 13 /// Dourado Precoce // IRAT 13PRA577 Latsidahy / CA 148PRA599 Pratao Precoce / Chhomrong DhanPRA620 Chhomrong Dhan / Lulu 281 = Chhomrong Dhan // Araguaia / CuiabanaPRA622 Chhomrong Dhan / SLIP 48-M-1 = Chhomrong Dhan // Chokoto / IRAT 263PRA632 CIRAD 391 / Luluwini10 = Latsidahy / Shin Ei // Lulu 291 / Lulu 183

91

ANNEXES

Annex 1.Abstracts

II Regional Agrociencia and Technology Seminar, Century XXI, ColombianOrinoquia. CORPOICA, PRONATTA. August 23 – 25, 2000. Colombia

andXXI Congress of Colombian Phitopathology Association and Similar Sciences(ASCOLFI):Pathology of post harvest in flowers, fruit bearing, vegetables, seeds, roots andtubers. Palmira, (CIAT), August 30 to September 1st, 2000, Colombia

Correlation Study Between Partial Blast Resistance (Magnaporthe Grisea)In The Field And In The Greenhouse

Dosman J.P. [jpaoladossmann@ postmark.net]. Vales M.,Tulande E., DuqueM.C. International Center of Tropical Agriculture, CIAT, Cali.

Rice blast, caused by Pyricularia grisea Sacc. (telemorfo Magnaporthe grisea(Hebert) Barr.) is a main disease of the crop all over the world. Resistantvarieties development is the most economic method to control the disease, beingthe best strategy to obtain a durable resistance, working with complete andpartial resistance genes. To establish in a safe way, a possible rice selectionwith partial resistance in a simpler way in greenhouse with respect to fieldevaluation, two fungus strains of lineages-6 and SRL4 were made, with previousknowledge of varieties behavior in field conditions with respect to these strains.Greenhouse evaluations were made from inoculations in 23 Latin America ricematerial that did not show complete resistance to these strains. Inoculationswere made by manual aspersion to 18 days old plants and then incubation ofmaterials started in plastic houses for 16 hours, after which those materials werewithdrawn and placed under greenhouse conditions for 8 days. After this period,plants evaluation starts taking into account mainly, types of lesion and leaf areaaffected. To study the correlation a statistic analysis was carried out with fieldresults previously with us and the ones found in the greenhouse. The analysisled us to conclude that it does not exist a correlation between the field andgreenhouse results. That means, that it is not exact to predict partial materialresistance with respect to an isolation in field, from data obtained in thegreenhouse.

Key words: Partial resistance, Pyricularia grisea, Rice3rd International CROP SCIENCE Congress 2000.

17-22 August 2000, CCH, Hamburg, Germany

92

Annex 2.

Breeding Strategy for Durable Rice (Oryza Sativa) Blast (MagnaportheGrisea) Resistance: Recurrent Selection for Complete, and PartialResistance, and for Other Agronomic Traits

M.Vales1, M.Chatel1, J.Borrero1, J.Dossmann1, E.Tulande2, M.Triana2, V.Kury2,M.C.Duque2, Y.Ospina1

1CIAT/CIRAD Collaborative Project, 2Rice Program, CIAT A. A. 6713 Cali,Colombia

Accumulation of complete and partial resistance genes is used to get a durableresistance against rice blast. Resistance polygenism leads to recurrentselection. Population PCT-6\HB has a male-sterile gene, which is resistant to Hoja Blancavirus (RHBV). We use thirteen strains from seven lineages of the cropping area.Every recurrent cycle includes three parallel parts with a common geneticrecombination:Selection for complete resistanceS1 lines presenting a broad complete resistance spectrum are selected in thegreenhouse, and are used for the next genetic recombination.In the field, during this recombination, a mass reciprocal plant-parasite selectionis done for complete resistance against natural fungus population.Selection for partial resistanceS2 lines are selected in field for partial leaf and neck blast resistance using acompatible strain. In the field, agronomic traits are also selected. Grain qualitiesand Togasodes orizicolus resistance are selected from S3 seeds and plants. Thebest S3 seeds are used in the next genetic recombination.Genetic polymorphism maintenance and male sterile gene frequencyIt is achieved by successive harvests of male sterile plants without selection. S0seeds are used for genetic recombinations.A cycle and half of the recurrent selection was carried out. Genetic progress forcomplete resistance is great. Assessment of genetic progress for partialresistance, and for agronomic traits is on the way. Successive samples ofenhanced population are used to improve varieties.

93

Annex 3.

International Symposium, Durable Disease Resistance. Wageningen, TheNetherlands, 11/28-12/1, 2000

Breeding Strategies For Durable Resistance To Rice (Oryza Sativa) Blast(Magnaporthe Grisea) Disease

Michel J. Vales

CIAT-CIRAD Collaborative project, Rice CIAT Project, CIAT, AA 6713, Cali, Colombia,[[email protected]]

Different breeding strategies (STR) are used to increase probability of gettingdurable resistance.

STR#1: Accumulation of complete resistance genes using lineagesknowledge. Although the risk of unknown compatible strains is here, a highescape was observed. Field and greenhouse evaluation, and plant-parasitereciprocal selection are proposed. This gene accumulation leads to the use ofrecurrent selection.

STR#2: Virulence incompatibility use. Selection of resistance genescorresponding to exceptional virulence genes association would allow a durablecomplete resistance. Recurrent populations with these resistance genes areformed.

STR#3: Use of complete resistance from other species. Rice genetictransformation would allow a durable complete resistance, but geneticallymodified organisms are very unpopular.

STR#4: General, polygenic, and partial resistance use:

General resistance (GR). Specific absence is never certain, and a generaladaptation is possible. Therefore for every selection cycle, strain inoculated inthe field is changed.

Polygenic resistance. Facts and theoretic models are in favor of polygenicresistance, though these models are unconvincing. Recurrent selection is theadequate method.

Partial resistance. We all know that complete resistance are specific, sopartial resistance is selected to discard this specificity. But part of partialresistance may be specific, so see GR. On the field, the difference between complete and partial resistance is notobvious because both reduce number and size of lesions. Compatible strain’ useis needed for partial resistance evaluation in the field

94

. Conclusion. A miraculous solution does not exist, so all available strategiesare proposed to decrease the risk of resistance breakdown.

Annex 4.

International Symposium, Durable Disease Resistance. Wageningen, TheNetherlands, 11/28-12/1, 2000

Durable Resistance to Rice (Oryza sativa) Blast (Magnaporthe grisea)Disease: Recurrent Selection Breeding Program of The CIRAD/CIATCollaborative Project

M.J. Vales1, M.Chatel1, J.Borrero1, J.Dossmann1, E.Tulande2, M.Triana2, V.Kury2,M.C.Duque2, and Y.Ospina1

1CIAT-CIRAD Collaborative Project, 2 CIAT, Rice Project, AA 6713, Cali, Colombia,[[email protected]]

Durability of resistance to rice blast disease is attempted by accumulatingcomplete and partial resistance. Resistance polygenism leads to recurrentselection breeding.

A rice population, with a male-sterility gene bred for the Hoja Blanca Virus(RHBV) resistance is used. Thirteen blast strains, from 7 lineages of the croppingarea, are used. Every recurrent cycle includes three parallel parts with a common geneticrecombination: - Selections for complete resistance. S1 lines showing a broad completeresistance spectrum are selected in the greenhouse and used in the next geneticrecombination. In the field, during this recombination, a reciprocal plant-parasiteselection is done for complete resistance against natural fungus population. - Selection for partial resistance. In the field, S2 lines are selected foragronomic traits and for partial leaf, and neck blast resistance using a compatiblestrain,. Grain quality, and resistance to Togasodes orizicolus are selected on S3seeds, and plants, respectively. Seeds from the best S2 lines are used for thenext recombination. - Genetic polymorphism, and male-sterile gene frequency maintenance. Toavoid a genetic drift and the lost of the male-sterility gene, successive harvests ofmale-sterile plants are done without selection. S0 seeds are used forrecombinations.

One cycle and half of recurrent selection was carried out. Genetic progress forcomplete resistance is spectacular. Assessment of genetic progress for partialresistance, and for agronomic traits is on the way. Successive samples ofenhanced population are used to improve varieties.

95

OUTPUT 2. CHARACTERIZING RICE PESTS AND THE GENETICS OF RESISTANCE

2.C. Characterization of the Complex of Rice Hoja Blanca Virus andTagosodes orizicolus

L.Calvert and R.Meneses

Objective: To Reduce Losses Caused by the Rice Hoja Blanca Virus(RHBV) and Tagosodes orizicolus and Determine the Impact on PestPrevention.

2.C.1. Introduction

Rice hoja blanca virus (RHBV) is member of the genus tenuivirus and is adisease of rice in the Caribbean, Central American and tropical AndeanCountries. The symptoms in rice are chlorotic streaks that can coealsce andcause the leaves to become white. When the young plant become infected, theyare stunted and in severe cases the leaves turn necrotic and the plants die.Infections that occur before the emergence of the panicle can reduce seed setand quality. It has been reported that RHBV infection predisposes rice toHelminthosporium oryzae and to brown spot of the grain.

RHBV epidemics are cyclic and the peaks of epidemics are separated by at leastseveral years. When RHBV is present, rice farmers tend to use multipleinsecticide applications to control the vector. These insecticide applicationsreduce the number of predators and parasites of T. orizicolus, and this can leadto higher than normal planthopper populations. T. orizicolus is a rice pest thatcauses direct damage both by feeding and laying its eggs on the plant. If theplanthopper becomes resistant to insecticides, the populations of theplanthoppers can reach very high levels and cause direct damage to the crop.

This research was carried out to prevent epidemics of RHBV and damagecaused by T.orizicolus. This report describes promising new varieties, progressin developing addition lines, as well as research to characterize resistance toRHBV and to T.orizicolus.

Outputs

2.C.2. Developing Materials Resistant to RHBV and Tagosodes orizicolus R.Meneses, L.Calvert, M.Triana, L.Reyes, and M.Cruz

2.C.2.1. Releasing Varieties Resistant to RHBV and its Vector

During the first semester of 2000, Colombia’s National Federation of Rice-Growers (FEDEARROZ) released two new rice varieties: Fedearroz 2000 andCOLOMBIA XXI. The potential yield of these varieties surpasses 6000 kg/ha, asshown in trials carried out in the central and in the dry Caribbean zones of the

96

country. The adoption of these varieties is expected to be widespread inColombia with a resulting decline in the area grown to Fedearroz 50. This varietycurrently planted in more than 60% of Colombia’s rice-growing area, but thereare risks in the stability of production when one variety is too dominate.

In experiments carried out at CIAT, variety COLOMBIA XXI (FB0100) showed anintermediate resistance reaction to RHBV. Fedearroz 2000 (CT10323-29-4-1-1T-2P) presented higher levels of resistance to the virus at 5 days afteremergence (DAE) than did the control resistant variety Colombia 1 or Fedearroz50. With the release of these materials, the control of the virus in commercialfields should be highly effective and farmers will have access to better options toeliminate the losses due to recurrent epidemics of RHBV.

Rice variety FUNDARROZ–PN1 was released this year in Venezuela. Seed ofthis variety, which is resistant to both T. orizicolus and RHBV, is currently beingmultiplied for mass introduction of the variety, basically in the rice-growing areanear the city of Calabozo in Venezuela.

All these varieties are highly resistant to the insect and to RHBV, which is basicto integrated crop management. Because of this collaborative effort betweenFLAR, Fundarroz, FEDEARROZ, and CIAT, farmers now have more options toreduce the losses caused by T.orizicolus and RHBV.

2.C.2.2. Maintaining Colonies of Tagosodes orizicolus and Testing RiceGermplasm for Resistance to RHBV and its Vector

Because of the increasing number of new lines developed by breeders, theprocess of maintaining and conserving colonies of both healthy and virulent T.orizicolus was optimized. This has allowed an increase in productivity as well asquality. These colonies are used for both greenhouse and field trials.

This report covers evaluations carried out during the second semester of 1999and the first semester of 2000. Tested lines are from CIAT, FLAR, Colombia(FEDEARROZ & CORPOICA), Venezuela (FUNDARROZ), Cuba (IIA), and Peru(table 2.16).

In open choice trials using mechanical damage as the criteria for resistance to T.orizicolus, 67% of the materials tested were from FLAR. This year there was anincrease in the number of lines from CIAT pre-breeding projects that were tested(1329) compared with 1999 (827) (table 2.17).

97

Table 2.16. Germplasm Tested at CIAT for its Reaction to RHBV (SemesterII 1999; Semester I 2000).

Score1SourceMaterials

testedR I S

FLAR VIOFLAR 310 62 62 186 F1 - Tropics (crossings) 304 234 0 70 F3 – Tropics 2,264 1,500 349 415 F4 – Tropics 2,186 1,144 294 748Germplasm Bank (BCF) 1,350 285 256 809FEDEARROZ- (Colombia) 3,115 962 573 1580FUNDARROZ (Venezuela) 275 99 61 115Countries (Cuba, Peru) 89 2 6 81Subtotal 9,893 4,288 1,601 4,004CIAT-Pre-Breeding 1,551 201 650 700CIAT-Pathology 1,000 0 0 1,000CIRAD-CIAT 557 208 210 139ICA-Colombia 152 16 18 118Total 13,153 4,713 2,479 5,961Percentage 100.0 35.8 18.9 45.3

1R = resistant (scale 1-3); I = intermediate (5); S = susceptible (7-9) (Rice standard evaluation system, IRRI, 1996).

Table 2.17. Reaction of Germplasm to Tagosodes orizicolus in Trials Heldat CIAT (Semester II 1998; Semester I 1999).

Score1Source Materialstested R I S

FLAR VIOFLAR 229 196 5 28 VIOAL 67 17 1 49 F4 –Tropics 2,142 963 145 1,034Germplasm Bank (BCF) 252 62 22 168FEDEARROZ (Colombia) 694 253 46 395FUNDARROZ (Venezuela) 327 91 24 212Countries Cuba (IIA) Peru

7611

66

25

680

Subtotal 3,798 1,594 250 1,954CIAT-Pre-breeding 1,329 351 52 926ICA (Colombia) 146 89 8 49CIRAD-CIAT 385 109 3 273Total 5658 2,143 313 3,202Percentage 100.0 37.9 5.5 56.6

1R = resistant (scale 1-3); I = intermediate (5); S = susceptible (7-9) (Rice standard evaluation system IRRI, 1996)

This year the number of lines assessed using mechanical damage by T.orizicolus was increased by 82%, basically because FLAR members used

98

Bluebonnet 50 as susceptible check, whereas the project methodology used IR-8as susceptible check. As a result, up to 14 days variation was observed in thefinal evaluation. The percentage of resistant lines was similar for 1999 (45%) and2000 (44%). The values obtained in the intermediate lines continue to be low,which can be attributed, among other causes, to the evaluation method used(table 3).

The same table also shows that, in evaluations of resistance to RHBV, a 37.2%reduction occurred as compared with 1999. The percentage of resistant linesdecreased from 60% (1999) to 36% (2000), probably because of the higherinsect pressure. Last year, the larger number of lines evaluated strained thecapacity of the viruliferous insect colony and there were few insects per plantreleased. This demonstrates the importance of carefully selecting the materialsfor evaluation and maintaining a limit to the number of materials evaluated ineach semester.

Table 2.18. Rice Lines Evaluated for Resistance to Tagosodes orizicolusunder Greenhouse Conditions and to The Hoja Blanca Virus in the Field(CIAT, 1997-2000).

Percentage2Pest1 Year

Total no.of lines R I S Retested

T. orizicolusT. orizicolusT. orizicolusT. orizicolus RHBVRHBVRHBVRHBV

19971998199920001997199819992000

1404323531065658

12,54214,76920,95213,153

4263454438476036

3786

11201919

3528475048332145

192003000

1 For effects of year, semester II of the previous year and semester I of the current year are considered. 2 R = resistant; I =intermediate; S = susceptible (Rice standard evaluation system, IRRI, 1996).

2.C.2.2.1. New Methodology to Assess the Mechanical Damage Caused by Tagosodes orizicolus

One of the main characteristics sought when developing new rice varieties inLatin America is resistance to mechanical damage T. orizicolus, encouraging thesearch for better ways to assess this damage.

The following three evaluation methodologies were compared in terms ofprecision: Weber (1987), Pantoja and Hernández (1992), and Weber modified(2000)(see references). The trial was carried out under greenhouse conditions at

99

CIAT, using the planthopper evaluation methodology that uses IR-8 assusceptible check. A total of 690 lines were evaluated with three replicates.

The percentage of resistant plants was 46.5% when the Pantoja and Hernández(1993) methodology was used, compared with 37,6% with the Weber (1987)methodology and 37.4% with the modified Weber (2000) methodology. Theselast two methodologies combine dead plants and visual evaluation (RiceStandard Evaluation System, IRRI, 1996). The information related to dead plantscan help breeders in future decision-making (table 2.19).

Table 2.19. Comparison of Three Methods to Evaluate Resistance toT.orizicolus

Damage Evaluation ScaleEvaluation method1 3 5 7 9

Pantoja (% of lines) 9.1 37.4 10.0 30.4 13.1Weber (% of lines) 8.3 29.3 11.9 36.5 14.0Modified Weber (% of lines) 8.3 29.1 11.9 29.4 21.3

The intermediate values, for the three methodologies, ranged between 10.0%and 11.9%, indicating that the probabilities of being ranked in this category aresimilar for the three. This, however, does not occur with resistant materials, thatreached higher values with the Pantoja and Hernández (1992) methodology. Theinverse occurred regarding susceptible materials for the other two methodologies(50.5%).

The stability of check varieties IR-8 (susceptible), CICA-8 (intermediate), andMakalioka (resistant) was evaluated. IR-8 and Makalioka were both found to bestable, with 100% susceptible and resistant plants, respectively. However, thevalues obtained by the intermediate check (CICA-8) varied broadly (table 2.20).

Table 2.20. Test for the Stability of the Weber Method of Evaluating forResistance to T.orizicolus

Damage evaluation scaleEvaluation of check varieties1 1 3 5 7 9

IR-8 (L-2) (% of lines) 0.0 0.0 0.0 8.0 92.0CICA-8 (% of lines) 0.0 22.0 24.0 32.0 22.0Makalioka (% of lines) 30.0 70.0 0.0 0.0 0.0

1IR-8 is susceptible, CICA-8 intermediate, and Makalioka resistant to T. orizicolus.

The Weber and modified Weber methodologies offer plant breeders greatersecurity in the selection of their materials because the probability of outliers in thecategory of mechanical damage by T. orizicolus is lower.

100

2.C.3. Identifying Possible Biotypes of Tagosodes orizicolus from DifferentRice-Growing Areas of Colombia

R.Meneses, L.Reyes, L.Calvert, M.Triana, M.Cuervo, and M.C.Duque


The damage caused by T. orizicolus and RHBV has increased in recent years,mainly in several Latin American countries, including Colombia. Farmers havebeen forced to apply different methods to control T. orizicolus, including theplanting of resistant varieties and the application of chemical insecticides.

Sogawa et al. (1984) put forward that, although resistance to the leafhopper hasbeen a major component in Integrated Management, this approach has beenthreatened in many countries because of the appearance of new insect biotypes.

The use of DNA molecular markers (PCR) has been incorporated into severalsystems and plays a very important role in the future improvement of plantspecies of agronomic interest, while contributing information on taxonomy,genetic variation, and evolutionary studies on arthropods (Williams et al., 1990).

To prove the existence of a new biotype, Saxena and Barrion (1985) argue that itis necessary to compile results of a series of studies dealing with, for example,differential varietal reactions, different responses to a single host, and theinsect’s morphological differences and cytological and enzymatic variations.

This study is evaluating the following aspects of three colonies of T. orizicolus:

The level of damage caused by feeding of T. orizicolus in different commercialvarieties: IR-8, Fedearroz 50, Oryzica Caribe 8, Oryzica 1, and Oryzica Llanos 5.The reaction to RHBV of Fedearroz 2000, Colombia XXI, Fedearroz 50,Colombia 1, and Bluebonnet 50.The differences in several biological parameters of T. orizicolus from the threedifferent colonies.The degree of control exerted by different insecticides.The evaluation of populations of T. orizicolus from several rice-growing areas ofColombia, using molecular techniques.

Also, all trials ultimately aimed to compile information to determine whetherbiotypes of T. orizicolus are present in the different rice-growing areas ofColombia. This study was carried out in laboratory, greenhouse, andscreenhouse conditions at CIAT, using colonies of T. orizicolus collected in ricegrowing areas of northern (Saldana) and southern Tolima (Lérida) and in Valledel Cauca (Jamundí). The insects came from rice fields that had not fumigatedwith insecticides. The insects were reared in 80 cm x 80 cm x 90 cm mesh cageson the rice variety Bluebonnet 50.

101

Five trials were conducted:2.C.3.2. Feeding of T. orizicolus on different rice varieties (mechanical damage)2.C.3.3. Evaluating varieties for resistance to RHBV2.C.3.4. Several biological aspects of the three colonies of T. orizicolus2.C.3.5. Efficiency of different insecticides in the control of adult T. orizicolus2.C.3.6. Evaluation of T. orizicolus using RAPDs

2.C.3.2. Feeding of Tagosodes orizicolus on Different Rice Varieties(Mechanical Damage)

Varietal resistance is one of the components of IPM. In recent years breedershave placed great emphasis on the incorporation of resistance to both pests anddiseases. Considerable advances have been made in developing germplasmresistant to T. orizicolus; however, the insect’s aggressiveness has not beenevaluated on commercial varieties with colonies from different rice growingregions.

To determine if there are differences in aggressiveness between the threecolonies of T. orizicolus, five rice varieties were infested from 5 and 10 days afterplant emergence, using a methodology similar to that used to determine themechanical damage caused by T. orizicolus.

The following varieties were evaluated: IR-8, Fedearroz 50, Oryzica Caribe 8,Oryzica 1, and Oryzica Llanos 5. These were planted at random in plastic traysmeasuring 50 cm x 25 cm x 8 cm. For each variety, 3 continuous rows of 20plants, spaced at 3 cm, were planted, with 4 replicates per colony. The trayswere transferred to the cages and infested with 1,200 non-viruliferous adult T.orizicolus (240 insects/variety).

The evaluation was carried out in the same way as the one used to evaluate themechanical damage caused by T. orizicolus when the susceptible check (IR-8) iscompletely affected. Weber’s evaluation method (1988) was followed in this trial.

The evaluations were carried out at 21, 23, 27, and 29 days after infestation(DAI) for both ages. Colonies and varieties were compared by Chi-squareanalysis.

Aggressiveness 5 days after emergence (DAE)The data obtained indicate that all varieties evaluated, except Fedearroz 50,present high plant mortality with the three colonies of T.orizicolus when infestedat 5 DAE.

The analysis of the aggressiveness of the three colonies revealed no differencesamong them, at all evaluation dates. As of 21 DAE, variety Fedearroz 50 showedhigh resistance, and continued to do so for all remaining evaluation dates (table2.21).

102

Table 2.21. Aggressiveness of Colonies of Tagosodes orizicolus in FiveRice Varieties (5 Days After Emergence)

Days after Infestation21 23 27

Variety NorthernTolima1

SouthernTolima

Valle NorthernTolima

SouthernTolima

Valle NorthernTolima

SouthernTolima

Valle

Fedearroz 50 R R R R R R R R ROryzica 1 S S S S S S S S SOryzica Llanos 5 S S S S S S S S SOryzica Caribe 8 S S S S S S S S SIR-8 S S S S S S S S S

1Colonies from different areas.

As it was previously pointed out, this test was not compared with others becauseit is the first of its kind that has been carried out in Colombia.

Aggressiveness at 10 days after emergence

Similarly to the evaluation at 5 DAE, no large differences were observed inaggressiveness of all colonies of T. orizicolus, although the colony from Valleturned out to be more aggressive on variety IR-8. Once again the varietyFedearroz 50, infested at 10 DAE, confirms its resistance to all three colonies(table 2.22).

Table 2.22. Aggressiveness of Colonies of Tagosodes orizicolus in FiveRice Varieties (10 Days After Emergence)

Days after Infestation21 23 27 29

VarietyN.

TolimaS.

Tolima ValleN.

TolimaS.

Tolima ValleN.

TolimaS.

Tolima ValleN.

TolimaS.

Tolima Valle

Federarroz50

R R R R R R R R R R R R

Oryzica 1 S S I S S S S S S S S SOryzicaLlanos 5

S S S S S S S S S S S S

OryzicaCaribe 8

S S S S S S S S I-S S S S

IR-8 I I I I S S I I S I I S

1Colonies from different areas.

When varieties were submitted to statistical analysis, Fedearroz 50 (FB 50)differed significantly from the other varieties, which did not differ amongthemselves. This variety showed high resistance to mechanical damage causedby the insect (Figure 2.4)

103

Figure 2.4. Evaluation of the Mechanical Damage Caused by Tagosodesorizicolus.

Although the colony from Valle del Cauca was less aggressive than the two fromTolima, no significant differences were observed between the three colonies.

No significant differences were observed in the aggressiveness of the threecolonies of T. orizicolus, indicating that this parameter cannot be used to detectthe presence of a new insect biotype.

All varieties were susceptible to the infestation at 5 DAE, except Fedearroz 50that showed resistance to all colonies.

2.C.3.3. Evaluating Varieties for Resistance to the Hoja Blanca Virus

Just as the mechanical damage caused by T. orizicolus is important, thedetermination of the effect of the origin of the Sogata colony on varietalresistance is essential to differentiating those colonies exerting greater pressureon rice varieties.

Adult T.orizicolus used were from the same colonies used to evaluate themechanical damage caused by the insect. Varieties were infested at 7 days afterplant emergence.

The following rice varieties were used: Fedearroz 2000, Colombia XXI,Fedearroz 50, Bluebonnet 50 (susceptible check), and Colombia 1 (resistant

104

check). These were planted at random in plastic trays measuring 50 cm x 25 cmx 8 cm. For each variety, 3 continuous rows of 15 plants, spaced at 3 cm, wereplanted, with 4 replicates per colony.

Prior to infestation, the virulence of each colony was determined by ELISA andon an individual plant basis. Results were as follows: the colony from Jamundí(78% and 86.6%); the colony from Saldaña (77%); and the colony from LaGuaira (79% and 88%). Trays were transferred to cages and infested with 900individuals of T. orizicolus per tray (4 insects/plant).

Insects were allowed to feed for 5 days on the materials, after which they wereeliminated by applying carbofuran (granules).

A randomized complete block design was used, and variance analysis andmultiple comparison tests were performed. Evaluations were carried out at 11,18, and 25 days after infestation (DAI), and the number of plants of each varietythat presented symptoms were counted.

Although no significant differences were observed between colonies, significantdifferences did occur between varieties. Bluebonnet 50 and Colombia XXI werehighly susceptible, with 99.4% and 85% of their plants presenting symptoms at25 days DAI (table 2.23).

Table 2.23. Percentage of Plants Presenting RHBV Symptoms

Days after Infestation (DAI)Variety Reaction toRHBV1 11 18 25

Fedearroz 2000 R 5.1 d 38.1 d 43.6 dColombia 1 R 15.7 c 62.8 c 66.1 cFedearroz 50 R 27.8 c 60.8 c 66.5 cColombia XXI I 34.0 b 80.9 b 85.0 bBluebonnet 50 S 86.6 a 96.3 a 99.4 a

1R = resistant; I = intermediate; S = susceptible (known reaction).

Variety Fedearroz 2000 differed significantly from the other varieties andsurpassed the resistant check Colombia 1 in terms of plant age when symptomsappeared and symptom severity.

No significant differences were observed between colonies, indicating that whencolonies from different localities are selected with similar percentages ofvirulence, the expression of symptoms of the hoja blanca virus depends on thevariety.

These results confirm that the resistance to RHBV of the new rice varietyFedearroz 2000 surpasses that of Colombia 1, considered the main donor ofresistance to this disease.

105

2.C.3.4. Several Biological Aspects of the Three Colonies of Tagosodesorizicolus

The study of insect biology is one of the major aspects of IPM because thisinformation can be used to design different control measures for the insect. Thistrial aims to determine several biological parameters of T. orizicolus, mainlythose related to increased pest population.

A total of 100 5th-instar nymphs were collected from each colony of T. orizicolusand those from each colony were placed in small cages with 25-day-old riceplants of the same variety. The nymphs were kept in these conditions until theyreached the adult stage. The following day 1 adult male and 1 adult female wereplaced on a rice plant of variety Bluebonnet 50, which had been covered with anacetate cylinder. Insects were left to feed on the plant for 4 days, after which thecouple was removed from that plant and placed on another plant, in similarconditions, and so forth until the female reached the end of its cycle.

Daily observations were made to record date of oviposition and emergence ofnymphs, and thus determine the length of nymphal stage. The total number ofemerged nymphs was counted and the total number per female was calculated.The length of the adult stage was also assessed. The length of the nymphalstage in the three colonies is quite similar for both females and males, rangingboth between 14.0 and 14.6 days (table 2.24)

In studies carried out in Cuba, Gómez and Kamara (1980) determined that thenymphal stage of T.orizicolus females averaged 14.9 days and that of males,14.3 days.

The average longevity of adult males from the colony of southern Tolima was33.6 days, a higher value than that reached by females from the same area, 30.8days, possibly because of the high oviposition of females from this colony, whichcan adversely affect their longevity.

In the other two colonies, females showed greater longevity than males: 33.0 forthe colony from Valle del Cauca and 31.8 days for that from northern Tolima. Theaverage difference among females from the three areas is 2.2 days (table 2.24).

The longevity of adult T. orizicolus was 24.12 days for males and 27.8 days forfemales (Rey and García, 1980).

106

Table 2.24. Length of Life Cycle of Tagosodes orizicolus from Colonies ofThree Rice-Growing Areas of Colombia.

Females (days) Males (days)LocalityNymphs Adults Nymphs Adults

Northern Tolima 14.6 31.8 14.6 27.6Southern Tolima 14.2 30.8 14.4 33.6Valle del Cauca 14.0 33.0 14.5 27.5

The total number of emerged nymphs per female is important because theprogeny of each female throughout its life cycle can be assessed. In the case offemale T. orizicolus from southern Tolima, the average emergence was 622.6nymphs/female, with a maximum of 827 in one female that lived 54 days. Thesevalues are much higher than those obtained in colonies from northern Tolima andValle del Cauca (table 10).

Table 2.25. Several Biological Parameters of Female Tagosodes orizicolus.

No. of emergednymphs/female

Days withoutdepositing eggs

before deathLocality

Averagelife cycle

(days)Average Max Min Max Min

Northern Tolima 53.4 360.6 565 (40) 231(18) 18 (50) 3 (32)Southern Tolima 56.0 622.6 827 (54) 429(46) 14 (46) 3 (25)Valle del Cauca 54.0 330.8 481 (40) 104(29) 17 (46) 3 (22)

( ) = Female longevity.

The average number of emerged nymphs/female in colonies from northernTolima and Valle del Cauca is lower than the minimum value reached in southernTolima: 429 nymphs/female T. orizicolus. This data is importance because theSogata population in that area increases are a rate nearly twice as fast as theother colonies.

Table 2.25 also indicates that, because of the longer adult stage, females stoplaying eggs earlier before dying. In the colony from northern Tolima a female witha 50-day cycle stopped ovipositing 18 days before dying compared with only 3days for female T. orizicolus with 25-32 day cycles in the three colonies.

No large differences were observed for both sexes in the length of the nymphalstage and life cycle of T.orizicolus of the three colonies.

Females of the colony from southern Tolima had a shorter longevity than malesof that same colony.

107

The total number of emerged nymphs/female was higher in the colony fromsouthern Tolima compared with the other two colonies.

2.C.3.5. Efficiency of Different Insecticides in Controlling Tagosodes orizicolus

The application of chemical insecticides is, undoubtedly, the most controversialof plant health activities. Agrochemicals have been—and will continue to be—powerful, but necessary, weapons to control insects. When designing an IPMstrategy it is important to determine how insecticides can be used more efficientlyto correct problems such as re-emergence and appearance of secondary pestsby using insecticides that are less toxic to natural enemies. In a large number ofIPM programs insecticides play a very importance role and their use can behighly profitable. However, their use is associated with many adverse secondaryfactors. As a result, in those cases in which chemical insecticides must beapplied, more selective insecticides, that perturb the rice agroecosystem less,should be selected.

This trial was carried out in semi-controlled screenhouse conditions to determinewhether the control of adult T. orizicolus by applying different insecticides wasinfluenced by the area in which the insects were collected, namely northern andsouthern Tolima and Valle del Cauca.

The relationship between percentage of mortality of T. orizicolus from the threecolonies and insecticide application was determined by Anova analysis and lethaltime with Probit analysis (1999 CIAT Annual Report).

Pots, 10 cm in diameter, were used, in which 3 plants of rice variety Bluebonnet50 were planted. Insecticides were applied 25 days after plant emergence, with aconstant pressure microsprayer (table 2.26).

Immediately after applying the insecticides, 20 adult female and male T.orizicolus were placed in each pot and the plants were covered with an acetatetube. Four replicates per treatment were used. The mortality of adult T. orizicoluswas assessed beginning 3 hours after insecticide application and continued untilmortality of insects was stable.

108

Table 2.26. Insecticides Applied to Control Tagosodes orizicolus.

Insecticide Type Formulation DoseEtofenprox C-H-O 10 EW 700 ml/haImidacloprid Chloronicotinyle 350 SC 100 ml/haCypermethrine Pyrethroid 200 EC 500 ml/haMonocrotophos Organophosphorus compound 600 EC 1000 l/haChlorpyrifos Organophosphorus compound 4 EC 1.0 l/haThiametoxan Nitroguanidine 25 WG 100 g/haAcetamiprid Chloronicotinyle 20 SP 150 g/haCheck

The control exerted by all insecticides on adults of the three colonies of T.orizicolus was higher than 92% at 168 hours after application.

In figure 2.5, imidacloprid illustrates the results of all insecticides. The controlexerted by this insecticide was similar for all three colonies of T. orizicolus. At 96hours after application, 81.9% control was reached in the colony from Valle delCauca; 85.0% in the colony from northern Tolima and 81.9% in the colony fromsouthern Tolima, confirming that no differences occurred in the control of adult T.orizicolus of the three colonies. This means that same doses and insecticidescan be used in all areas where T.orizicolus was collected.

In addition to the percentage of control, the Lethal Time (LT) 50 and 90 is keyinformation to determine in time that an insecticide takes to reach thesepercentages of mortality. These values differed for each insecticide, beingoutstanding in the case of monocrotophos. Data, however, were similar for allthree colonies.

The percentage of control and LT 50 and 90 change depending on theinsecticide, but the three colonies respond consistently to all treatments.

109

Figure 2.5. Adult mortality of three colonies of Tagosodes oriziculus treated withimidacloprid in greenhouse conditions (CIAT 1999 B).

0

20

40

60

80

100

0 24 48 72 96 120 144 168Hours after application

Perc

enta

ge m

orta

lity

Valle del Cauca (obs) Northern Tolima (obs) Southern Tolima (obs)Valle del Cauca (est) Northern Tolima (est) Southern Tolima (est)

2.C.3.6. Evaluation of Tagosodes orizicolus using RAPDs.

Other techniques, such as random amplified polymorphic DNA (RAPD), havebeen used to differentiate and identify homopteran populations and species suchas Aedes aegyptis. However, no reports have been published on the use ofmolecular markers in T. orizicolus, nor any similar type of work that could helpdifferentiate insect populations in different areas to establish possible biotypes ofthe insect.

The identification of molecular markers (RAPD/PCR) in the DNA of insects allowsspecies to be differentiated among themselves. Sixty primers of Operon, OPO,OPN, OPL, OPM, and OPP (Operon Technologies, INC, California, USA) kits,available at CIAT’s Virology Laboratory, were evaluated in insects gathered inother rice-growing areas such as Jamundí (Valle del Cauca), Huila (centralColombia), and Magdalena (Colombia’s northern coast). Another type ofTagosodes was included: T. cubanus as differentiation pattern with T. orizicolus.

Of the 40 primers evaluated with insects gathered from the rice-growing areas ofJamundí and southern Tolima, the primer OPO 5 (CCCAGTCACT) showed aPCR product that was present most of the time in one of the populations thatcould help establish differences between the 2 populations. The RAPDs did notclearly different between populations of T. orizicolus.

With the primer OPN 5 (ACTGAACGCC), T. cubanus showed differences as canbe observed in Figure 3, as compared with T. orizicolus . This confirms thepresence of two different species.

There were consistent difference with several primers and the species T.orizicolus and T. cubanus can be differentiated using molecular markers.

110

Figure 2.6. Primer OPN 5 (ACTGAACGCC). No. 1 to 10: Tagosodesorizicolus. No. 11 and 12: Tagosodes cubanus. * = Marker of molecularweight 1 Kb.

Conclusions

1. Based on the trials carried out in this study, none of the Colombian coloniesstudied represent a distinct biotype of T. orizicolus.

2. A general control strategy for T. orizicolus can be designed for the threeregions because there are not differences among them.

2.C.4. Interactions of RHBV and T. orizicolusL.Calvert, R.Meneses, L.Reyes, M.Triana, M.Cruz, C.Pardey and M.Cuervo.

2.C.4.1. Monitoring Rice Growing Zones for RHBV and T.orizicolus

Since 1996, FEDEARROZ, CORPOICA and CIAT, with the support ofColciencias, have collaborated to monitor major rice growing regions inColombia. This activity was extremely useful in identifying those areas that wereat great risk of outbreaks of RHBV. This has allowed us to concentrate ourefforts on those areas in greater need.

During the last two years Fedearroz 50 has become the predominant rice varietygrown in Colombia. Rice producers also increased the area with several rice

111

varieties such as Selecta 320, Orizica 1, Orizica 3 that have an intermediate levelof field resistance to RHBV. With changes in varieties and good managementpractices, especially in pesticides’ use, levels of RHBV have dramaticallydecreased. During this year, intensive monitoring was greatly reduced. Insteadof monitoring all zones, in a somewhat random fashion, we are only surveyingthose areas in which RHBV is reported to be present in at least moderate levels.This year, there were very few reports of RHBV and our collaborative effortshave contributed to rice growers by reducing losses caused by the disease andcosts to manage the problem.

2.C.4.2. Interactions of Host/Vector/Virus

2.C.4.2.1. Trials for RHBV Field Resistance

All experiments were done using a randomized plot design of three blocks withtwo replications. Each replication was a 1m2 plot of ten rows each planted with1.5g of seeds. Disease pressure was regulated using three levels of infestationwith T.orizicolus. Ninety T.orizicolus were tested using ELISA specific for RHBVto estimate level of virulence in the colony. Estimate for high level of infestationwas 3 plant hoppers per plant. Low infestation level was approximately 1 planthopper per plant. No insects were released in control plots. Plant hoppers werereleased 23 days after planting. To minimize the effect of movement of insects,the experiment was arranged with low infestation block placed in between thecontrol and the high infestation blocks. Rice barriers were placed around eachblock. To determine the level of RHBV, a visual rating of each row was madeusing evaluation scale of 0 (no symptoms), 1 (less than 1% of the plants withsymptoms), 3 (1 to 10%), 5 (11 to 30%), 7 (31-60%) and 9 (61-100%).

In the experiment there were five commercial varieties, Cica 8, Oryzica 1,Selecta 3-20, Fedearroz 50, Línea 2, and five advanced lines CT10323-29-4-1-1T-2P, FB100-10-1-M-1-M, P5419-2-20-11-1B, CT10192-5-1-2-2T-1-1, andCT10240-10-1-2-IT-2-1, from Colombia; three commercial varieties, INIA 14,Capirona, and Uquihua, and two advanced lines PNA-2002-HUA-2-EP2-1-PH3,and CT10310-15-1M-YA1-EP1 from Peru; and two commercial varietiesCimarrón and PNA97004, from Venezuela.

Susceptible check Cica 8 had a rating of 8.9 and a yield of 0.2 t/ha in the highinfestation block; a rating of 8.7 and a yield of 1.3 t/ha in the low infestation block;and a rating of 0.3 and a yield of 6.3 t/ha in non infested block with virulent T.orizicolus. This was the indicator that the experiment had very high inoculumpressure.

Most of the varieties were susceptible to RHBV in the high infestation treatment.This was not unexpected since resistance is only partially expressed 15 daysafter planting, when vectors were released in the field. Only two of the new linesfrom Colombia, CT10323 and CT 10240, had a rating of less than three (table

112

2.27). Ratings of 1 and 3 are considered to be highly resistant to RHBV. Thenext best line was the new variety recently released in Venezuela, PN97004, witha rating of 4.5 in high infestation. Since this performed significantly better, inseveral trials, than Fedearroz, 50, which had a rating of 6.6 in the highinfestation, this variety should perform well in field conditions, as highly resistantto RHBV. The best Peruvian line Peru was CT10310 with a rating of 6.2. Thisvariety performs similarly to Fedearroz 50 and if it becomes a variety, it will be animprovement over existing varieties. Line PNA-2002 was highly susceptible toRHBV.

Table 2.27. Field evaluation of rice hoja blanca virus using different T.orizicolus infestation levels 23 days after planting.

Level of infestation with T. orizicolus1

High infestation Low infestation No infestationRice lines Level of

RHBV2YieldT/ha

Level ofRHBV

Yieldt/ha

Level ofRHBV

Yieldt/ha

CT10323 1.4 jkl3 3.9 ghijkl 1.0 l 6.7 a4 0.0 l 6.5 a

CT10240 2.3 jk 3.7hijklm 1.1 kl 6.2 abcd 0.1 l 4.6 efghij

CT10310 6.2 ef 2.6lmnop 4.5 gh 4.8 cdefghi 0.0 l 5.0 bcdefgh

Fedearroz 50 6.6 def 1.7nopqrs 3.9 h 4.7 defghi 0.1 l 6.6 a

FB100 5.9 ef 1.6opqrs 4.2 gh 5.3 abcdefg 0.1 l 4.0 fghijkl

Selecta 3-20 7.0 cde 1.6opqrs 6.5 def 3.2 jklmn 0.2 l 5.5 abcdef

P5419 6.4 def 1.4opqrs 2.7 ij 2.7 lmnop 0.1 l 4.6 efghij

Inia 14 8.2 abc 1.3 pqrs 8.3 abc 2.3 mnopqr 0.2 l 6.4 abUquihua 7.2 bcde 1.3 pqrs 5.3 fg 4.2 fghijk 0.1 l 6.5 aCapirona 8.1 abc 1.2 pqrs 6.6 def 2.8 klmno 0.1 l 6.0 abcdePNA97004 4.5 gh 1.2 pqrs 2.5 j 4.8 cdefghi 0.1 l 4.3 fghijkOryzica 1 7.6 abcd 0.9 qrs 3.8 hi 4.0 fghijkl 0.1 l 3.3 ijklmLínea 2 7.0 cde 0.8 rs 5.9 ef 3.6 hijklm 0.1 l 4.3 fghijkCimarrón 8.6 ab 0.6 s 7.7 abcd 2.4 mnopq 0.1 l 5.5 abcdefPNA-2002 8.6 ab 0.3 s 6.0 ef 1.7 nopqrs 0.0 l 5.5 abcdefCica 8 8.9 a 0.2 s 8.7 a 1.3 pqrs 0.3 l 6.3 abc

1Infestation was approximately 3, 1 and 0 insects per plant.2A Visual scale of 0-9 was used to rate level of RHBV infection.3Averages for RHBV level with the same letter show no significant differences at the 5% level in Duncan multiple range.4Averages for yield t/ha with the same letter show no significant differences at the 5% level in Duncan multiple range.

In the commercial fields, Fedearroz 50 has proven to be a RHBV resistantvariety. Using the two phase screening for RHBV for the development ofresistant varieties is proving to be highly successful. Two recently releasedvarieties, FB100 and PN97004, as well as four of the new lines, were equal or

113

better than Fedearroz 50. If this trend continues, farmers will have a wide rangeof choices in those rice growing regions where RHBV is a cyclic problem.

Although potential new rice lines from three countries were represented in thecurrent field trial, these are still countries without adequate varieties were RHBVis a recurrent problem. Given the need to test more materials, this type of trialwill be modified to allow a greater number of varieties to be screened.

2.C.4.2.2. Testing for plant reactions to mechanical damage by T.orizicolus.

The varieties IR 8 (susceptible control), Makalioka (antibiosis control), O. llanos5, Fedearroz 50 and its parents P1274 and O. llanos 4 were compared todetermine more accurately the level of mechanical damage to T. orizicolus whenthe insects were given a choice on which variety to feed. Rice lines were grownin rows each containing 10 plants and were evaluated for mortality when thesusceptible control IR8 had more than 80% mortality in every row.

In these open choice experiments O. llanos 5 is equally susceptible tomechanical damage as IR 8. Llanos 5 has resistance to RHBV but not to T.orizicolus, and when the invasion of insects is early in the growing cycle, the fieldreaction of Llanos 5 is similar to susceptible varieties. This is further evidencethat the lack of resistance to the insect is an important component to resistanceto RHBV. In the second group the line P1274-6-8M-1-3M-1 had an intermediatelevel of mechanical damage by T. orizicolus and O. llanos 4 has a low level ofmechanical damage. Fedearroz 50 had the same low level of mechanicaldamage as O. llanos 4.

This test measures the mechanical damage and not the tolerance to the insect. Aseries of experiments were made to calculate the indexes of tolerance to theinsect, but the other mechanisms of resistance make it difficult to interpret theresults and to conclude the level of tolerance of Fedearroz 50 to T. orizicolus. Itappears that the low level of mechanical damage is due to mechanisms ofresistance other than tolerance.

2.C.4.2.3. Testing for antibiosis in Fedearroz 50.

Last year, the settling and ovipositional preferences of T. orizicolus on these sixrice lines were reported. Studies on the life cycle of T. orizicolus were made bycomparing mortality, duration, sex ratios of survivors, the weights of adults,nymphs, and the plants as well as other factors on the six varieties in this study.A further study was made to determine prepare life tables for IR 8 and Fedearroz50. The weight of adults grown on each of the six varieties is shown in table 2.28.IR 8, P1274 and O. llanos 5 had similar weights. The average weight of theinsects reared on Fedearroz 50, O. llanos 4 and Makalioka were similar to eachother but less than half the weight of the insects reared on the more susceptible

114

varieties. The major difference between the three lines was the high mortality ofinsects on Makalioka. The results of this series of experiments indicate that thevariety Fedearroz 50 has both antixinosis and antibiosis as mechanisms ofresistance to T. orizicolus. Fedearroz 50 is a very popular variety and theantibiosis may put pressure on T. orizicolus and create new biotypes.

Table 2.28. The weight of T. orizicolus reared on six varieties of rice.

VarietyNumber

offemales

Averagefemaleweight(10-4 g)

Numberof males

Averagemale weight

(10-4 g)

IR 8 27 6.26b1 32 2.75aP1274 29 8.31a 29 2.48aO. llanos 5 31 7.94a 28 2.71aO. llanos 4 27 3.15c 33 2.70aFedearroz 50 31 2.10c 20 1.80bMakalioka 3 2.172 2 2.5021Averages for the weight with the same letter show no significant differences at the 5% level in Duncan multiple range.The analysis of the males and females were made separately.2The results of Makalioka could not be compared statistically because of the low number of survivors.

2.C.4.2.4. Using adult T. orizicolus for evaluation of RHBV resistance infield screening trials

In the field trials, the majority of the insects released are at the nymphal stage. Inthis experiment, the majority of the insects released were at the adult stage.Design of experiment was similar to that described in section 2.3.2.1, with theexception that the release of insects was 33 days after planting, and that insectswere adult T. orizicolus. Level of infestation was approximately 1.4 insects perplant. RHBV levels were moderated with CICA 8, the susceptible control rated as7.7. The best line was CT 10323, and it was rated as 1.3 (table 2.29). Otherlines were rated between 3 and 4.5. The surprising result was the difference inyield. Losses were much higher than expected for the levels of RHBV infection.This is the first time that this experiment is carried out, and environmentalconditions were unusually rainy with low light intensity. Plots seemed to be freeof biological factors that could have affected the yield. It is well known that T.orizicolus adults cause more damage and are more active than nymphs. Still thisexperiment must be repeated before concluding that adults, in general, cause amore severe problem than nymphs. If these results are repeatable, then anevaluation of the field methodology is needed, primary RHBV infection is bymigratory insects.

115

Table 2.29. Comparison of reaction of 6 varieties to a T. orizicolusinfestation 33 days after planting.

No infestation Infested ca.1.4 adults/plant

Varieties Incidence of RHBV1 Incidence of RHBV

Capirona 0.1 4.5Fedearroz50 0.1 3.5

CT10323 0.0 1.3Cica 8 0.3 7.7PNA97004 0.1 3.1Oryzica 1 0.1 4.5

1A visual scale of 0-9 was used to rate level of RHBV infection.

Understanding components of RHBV resistance and its vector T. orizicolusThis year, there was an increased emphasis in understanding mechanisms ofresistance to RHBV and its vector T. orizicolus. These series of experiments willtake several years and will form the basis for marking genes for resistance toRHBV and T. orizicolus. Seven varieties (HB core varieties) were chosen with arange of resistance to the virus and the vector. Since resistance to RHBV isrelative, variety Colombia 1 which is one of the principal source of resistance toRHBV, is called resistant (R). Orizica 1 and Llanos 5 are designated asintermediate resistant (IR), because they are more susceptible, in our tests, thanColombia 1. The variety Fedearroz 2000 is apparently more resistant thanColombia 1, and is called highly resistant (HR). One must be cautious in definingresistance RHBV because it is not well understood how components ofresistance to the vector contribute to virus resistance. Since T. orizicolus mustbe used to inoculate rice with RHBV, relationship between vector and virus isdifficult to separate. It is easier to determine which is the plant resistance to T.orizicolus, because one can use virus-free plants. Resistance to T. orizicolusreported on table 2.30 is to mechanical damage in a test where many lines areavailable. Understanding the interaction of these different types of resistance isan objective of this research.

116

Table 2.30. Reaction to RHBV and T. orizicolus of varieties being used todetermine the components of resistance to RHBV and T. orizicolus

ReactionVariety RHBV Sogata

IR 8 S SCica 8 S IROryzica Llanos 5 IR SOryzica 1 IR RColombia 1 R SFedearroz 50 R RFedearroz 2000 HR R

2.C.4.3. Interactions between time of infestation and disease pressure.

Main source of resistance to RHBV has been Colombia 1. This variety is moresusceptible when it is young and becomes more resistant with age. Thirty daysafter planting, plants become almost immune to RHBV. There is also a dosageeffect. The more inoculum pressure in form of virulent plant hoppers, the greaterthe percentage of plants that become infected with RHBV. The HB core varietieswere tested in the field at two different levels of infestation 17 and 26 days afterplanting. The design was the same one described for testing variety in section

2.C.4.3.1. Rated for RHBV incidence

Results correlate well with predicted reactions based upon interactions of RHBVresistance and T. orizicolus (table 2.31). Both IR8 and Cica 8 reacted as highlysusceptible varieties and were highly infected with RHBV. Only at the last lowlevel infestation, did these varieties have a moderate level of incidence (IR 8; 5.9vs. Cica 8; 4.2), and this difference may be attributed to intermediate resistancelevel to plant hopper in Cica 8. Expression of virus resistance is most notable inRHBV incidence reduction 26 days after planting as compared with 17 days afterplanting. Incidence of RHBV at 26 days correlates well with virus resistance’srating of each variety. This is less true at 17 days, but this is expected since virusresistance is only partially expressed at that age. The exception is Fedearroz2000, which was highly resistant in all treatments. This is a further evidence thatFedearroz 2000 is a variety which is highly resistant to RHBV infection.

117

Table 2.31. Effects of time and number of virulent insects on RHBVincidence

Infestation 26 daysafter planting

Infestation 17 daysAfter planting

Variety High1

InfestationLow

InfestationHigh

InfestationLow

InfestationNo

infestation

IR 8 9.0a2 5.9ef 8.6a 8.4ab 1.0jkCica 8 8.9a 4.2gh 8.9a 8.9a 1.0jkO. Llanos 5 8.6a 5.0fg 8.0abc 7.3bcd 1.2jkOryzica 1 6.1ef 2.9i 7.2cd 5.7ef 0.8jkColombia 1 6.3de 1.8j 7.9abc 5.8ef 0.6jkFedearroz 50 3.9hi 1.3jk 8.2abc 6.3de 0.5jkFedearroz 2000 1.0jk 1.0jk 1.1jk 1.7j 0.2k1A visual scale of 0-9 was used to rate the level of infection with RHBV.2Averages for RHBV level with the same letter show no significant differences at the 5% level in Duncan multiple range.

2.C.4.4. Testing for tolerance to Rice Hoja Blanca Virus

One of the potential mechanisms for resistance is tolerance to RHBV. Toleranceis defined as the ability of the plant to sustain effects of a disease without dyingor suffering serious injury or crop loss. To determine if a plant is tolerant, it mustbe infected with the pathogen. Three blocks with two repetitions containing threerows was the basic experimental design. To assure all the plants were infectedwith RHBV, one gram of seeds were plant per row and this was rouged to tenplants, which showed symptoms. Negative controls were rouged ten plantswhich did not display symptoms of RSNV. There were three level of infestationcorresponding to approximately 6, 2, 0 plant hoppers per plant. Plots wereinfested at 23 days after planting.

At high levels of inoculum pressure, the three varieties tested are not tolerant toRHBV (table 2.32). In comparison to negative controls, yield losses were morethan 85%. At the lower level of inoculum pressure the yield losses were morethan 60% for the susceptible variety O. Caribe 8 and for the resistant varietyFedearroz 50. Oryzica 1 was the only variety that may have some tolerance, avariety which only had a decrease in yield of 38% in the lower level of inoculumpressure. This experiment will be repeated using all seven HB core varieties.Preliminary evidence is that tolerance is not an important mechanism forresistance to RHBV.

118

Table 2.32. Tolerance and the effect of dosage of RHBV.

High infestation6 nymphs/plant

Low infestation2 nymphs/plant

Noinfestation

Variety weight(g)2

% loss weight(g)2

%loss

weight(g)

O. Caribe 8 26.3 87.2 69.2 66.2 205.0Oryzica 1 28.0 84.8 114.4 37.8 184.0Fedearroz 25.0 91.6 114.0 61.8 298.8

2.C.4.5. Determination of the level of damage caused by Tagosodesorizicolus

2.C.4.5.1. Introduction

Rice variety Fedearroz 50 is planted in approximately 60% of Colombia’s rice-growing areas, and the percentage of virulence of Tagosodes orizicolus differs inthese areas. It is therefore important to determine yield losses caused bydifferent levels of virulence to offer the farmer IPM alternatives to control thepest.

2.C.4.5.2. Materials and Methods

Variety Fedearroz 50 was planted at a rate of 3 g/row in plots 1.05 m x 1.20 m,with 4 replicates per treatment (7 rows, spaced at 15 cm).

Nylon cages were placed in plots at 7 and 15 days after emergence beinginoculated with adult T. orizicolus according to each treatment. Insects fed onrice plants for 5 days, after which adult T. orizicolus were removed and aninsecticide applied to keep plots insect-free (table 2.33).

119

Table 2.33. Total number of adult Tagosodes orizicolus

Total number of adult T. orizicolusPer plot Whole treatment

Treatments Healthy Virulent Healthy Virulent0.25 insect/plant (5%) 43 2 172 80.25 insect/plant (10%) 41 4 164 160.25 insect/plant (healthy) 45 0 180 00.50 insect /plant (5%) 86 4 344 160.50 insect/plant (10%) 81 9 324 360.50 insect /plant (healthy) 90 0 360 0Non-infested check 0 0 0 0Total 386 19 1,544 76

The presence of RHBV was assessed 45 days after rice plants were infested.The crop was left until harvest to determine yield. Analysis of variance andmultiple comparison tests were performed.

2.C.4.5.3. Results and Discussion

In plots infested at 7 days after emergence (DAE), the treatments involving 0.25insect/plant with 10% virulent insects, 0.50 insect/plant with 5% virulent insects,and 0.50 insect/plant with 10% virulent insects did not differ significantly (table2.34).

Table 19 indicates that, with infection at 15 DAE, plants did not show symptomsof RHBV and no significant differences in yield were observed betweentreatments.

120

Table 2.34. Percentage of RHBV and yields of rice variety Fedearroz 50,under different Tagosodes orizicolus populations and percentages ofvirulence.

Percentage of RHBV and YieldTreatments 7 DAE

(%)Yield

(kg/ha)15 DAE

(%)Yield

(kg/ha)0.25 insect/plant (5%) 0.4 b 6,720 0.0 6,7210.25 insect/plant (10%) 2.9 a 7,595 0.0 6,6440.25 insect/plant (healthy) 0.0 b 7,653 0.0 6,4570.50 insect/plant (5%) 2.5 a 6,802 0.0 6,7110.50 insect/plant (10%) 2.1 a 7,280 0.0 6,9760.50 insect/plant (healthy) 0.0 b 7,741 0.0 6,632Non-infested check 0.0 b 7,750 0.0 6,817

No significant differences occurred in yield among the different treatments at bothinfestation times, indicating that the levels of insect infestation were very low andthat the symptoms presented did not affect rice yields (table 2.34).

The highest percentage of RHBV infection reached was 2.9%, a value equivalentto 1 of IRRI’s Rice Standard Evaluation System (1996). This low percentage ofplants presenting RHBV symptoms could be attributed to the number of insects(virulent) used to infest the crop.

Subsequent trials should increase the insect population/plant to be able todetermine differences among treatments.

2.C.5. Fluctuation of Tagosodes orizicolus populations in different ricevarieties


One of the limitations of rice cultivation in Colombia is the pest Tagosodesorizicolus, which causes mechanical damage to the crop and serves as vector ofthe hoja blanca virus. The insect is usually controlled with insecticides; in most ofthe cases, varietal resistance is not taken into account. The present study aimsto quantify the populations of T. orizicolus and its natural enemies in three ricevarieties planted in a commercial lot, with no application of insecticide.

2.C.5.2. Materials and Methods

The trial was carried out at the Villa Alejandra farm, located in the rice-growingarea of Guacarí (Valle del Cauca). Rice varieties Fedearroz 50, Oryzica Llanos 5,and Oryzica Caribe 8 were planted at a rate of 180 kg seed/ha. Soil preparation,

121

weed control, and fertilization were performed according to local technology.Insects were collected beginning 8 days after planting up to 102 days afterplanting, at 8-day intervals. Three samples were taken per variety per visit andeach sample consisted of 10 double passes of the entomological net. Varietieswere left until harvest and yields were determined. Data were submitted tovariance analysis and multiple comparison tests.

As indicated in table 2.35, significant differences occurred among varietiesregarding populations of T. orizicolus females, nymphs, adults (females + males),and total number of insects (females + males + nymphs).

Table 2.35. Populations of Tagosodes orizicolus in three rice varieties(Guacarí, Valle del Cauca, 2000).

Tagosodes orizicolusVariety Females1 Nymphs1 Total1

Oryzica Llanos 5 20.6 a2 38.2 a 74.1 a Oryzica Caribe 8 17.4 b 26.0 a 60.9 a Fedearroz 50 10.2 c 14.1 b 35.5 b

1Values presented correspond to the average of 13 samplings, each with three replicates.2Multiple comparison tests were performed by means of Duncan (p = 0.05).Treatments with the same letter did not differ significantly.

The T. orizicolus population in rice variety Fedearroz 50 differed significantly fromthat of the other varieties, presenting the lowest number of insects (85 adults/10double passes of the entomological net). The highest insect population wasreached in Oryzica Llanos 5 at 84 DAE (130 adults/10 double passes of theentomological net) and in Oryzica Caribe 8 at 98 DAE (215 nymphs/10 doublepasses of the entomological net) (Figure 2.7).

122

-5

10

25

40

55

70

85

100

115

130

14 28 42 56 70 84 98 112

Days after emergence

Inse

cts/

10 d

oubl

e pa

sses

of n

et

Fedearroz 50 O. Llanos 5 O. Caribe 8

Figure 2.7. Influence of rice variety and age on population ofTagosodes orizicolus (Guacari, Valle del Cauca, 2000).

Adults Nymphs

-10

5

20

35

50

65

80

95

110

125

140

14 28 42 56 70 84 98 112

Days after emergence

Inse

cts/

10 d

oubl

e pa

sses

of n

etFedearroz 50 O. Llanos 5 O. Caribe 8

The low presence of T. orizicolus in Fedearroz 50 confirms results previouslyobtained (fluctuation of insect population, Annual Report, 1999).

The highest population of T. orizicolus occurred in variety Oryzica Llanos 5,indicating that the insects prefers feeding on this variety compared to Fedearroz50 and Oryzica Caribe 8.

No significant differences were found in percentage of parasitism or in spiderpopulation among the three varieties.

2.C.5.3. Conclusions and Recommendations

The lowest values of T. orizicolus, in all stages of development, were reached invariety Fedearroz 50.Because of the characteristics of this variety, the application of pesticides shouldbe restricted during the crop cycle.

Bibliography

GÓMEZ, J . and F. KAMARA. 1980. Determinación de algunos parámetros deSogatodes orizicola (Muir). Centro Agrícola (Cuba)7(3).

REY XIOMARA and A. GARCÍA. 1980. Estudios de algunos aspectos del ciclobiológico del Sogatodes orizicola. Cienc. Tec. Agri. Arroz. 3(1).

SAXENA, R. and A. BARRION. 1985. Biotypes of the brown planthopperNilaparvata lugens (Stal.) and strategies in development of host plant resistence.Insect Sci. Applic. 6(3):271-289.

123

SOGAWA, K., K. FJSYNIKS , and H. BLAGIAWATI. 1984. Characterization ofthe brown planthopper population on IR-42 in North Sumatra, Indonesia. IRRN9(1):25.

WEBER, G. 1988. Metodología de trabajo en entomología de arroz. Programade Arroz. CIAT. 51 p.

Williams, J. G. K., Kubelit, A. R., Livak, K. J., Rafalki, J. A., and Tingey, S. V.(1990). DNA polymorphism amplified by arbitrary are useful as genetic markers.Nucleic Acids Res. 18: 6531-6535.

2.C.6. Training

Different meetings, courses, and other training events on IPM, particularlyregarding T. orizicolus and the hoja blanca virus, have been held with theparticipation of farmers and agronomists.

2.C.6.1. Conferences

Date Locality Lecturer TopicNovember 1999 Cúcuta L. Reyes RHBV-Tagosodes orizicolus and

RSNVDecember 1999 Geneva L. Reyes RHBV-Tagosodes orizicolus and

RSNVFebruary 2000 Jamundí L. Reyes RHBV-Tagosodes orizicolus and

RSNVMarch 2000 Ibagué M. Triana Tagosodes orizicolus /RHBVMay 2000 CIAT M. Triana RHBVSeptember 2000 CIAT M. Triana

Tagosodes orizicolus /RHBVSeptember 2000 CIAT1 M. Triana Breeding for resistance to insects and

evaluation of RHBVSeptember 2000 CIAT1

L. ReyesIntegrated management of Tagosodesorizicolus and selection of advancedlines for resistance to RHBV

September 2000 CIAT1 R. Meneses IPM1. Course on Application of Conventional and Molecular Methods to Rice Improvement, held at CIAT, 24 September-6

October 2000.

2.C.6.2. Training offered

Date Person and Country Topic2 June-24 July2000

Miguel Muñoz (Costa Rica) Management of Tagosodesorizicolus/RHBV

124

2.E.7.3. Participation in scientific events

Date Name Event Activity13-14March/2000

Rafael Meneses Workshop on information systemfor agroecological crop production.Ibarra, Ecuador.

Lecturer

26-28JULY/2000

MÓNICA TRIANA XVII CONGRESS OF THE COLOMBIANENTOMOLOGY SOCIETY, MEDELLÍN

LECTURER

26-28JULY/2000

LUIS A. REYES XVII CONGRESS OF THE COLOMBIANENTOMOLOGY SOCIETY, MEDELLÍN

LECTURER

26-28JULY/2000

MARIBEL CRUZ XVII CONGRESS OF THE COLOMBIANENTOMOLOGY SOCIETY, MEDELLÍN

LECTURER

Publications

Evaluación de insecticidas para el control de Tagosodes orizicolus encondiciones de campo, I parte. Lee Calvert; Rafael Meneses; Luis AntonioReyes; Alexander Pérez. CORREO. Fedearroz, Noviembre de 1999. No. 107.pp. 4-5.

Evaluación de insecticidas para el control de Tagosodes orizicolus encondiciones de campo, II parte. Lee Calvert; Rafael Meneses; Luis AntonioReyes; Alexander Pérez. CORREO. Fedearroz, Diciembre de 1999. No. 108.pp. 4-5.

Estrategias para el control del virus de la hoja blanca en Colombia. L. Calvert yL. Reyes. Final report submitted to Colciencias. Monograph. 35 p. March 2000.

Encuesta en el manejo de insectos plaga del arroz en Colombia. Luis AntonioReyes; Lee Calvert. CORREO. Fedearroz, Abril del 2000. No. 112.pp. 4-5.

Collaborators

Efren CórdobaMauricio MoralesRodrigo MoránJames SilvaJulio Holguín

125

OUTPUT 2. CAHRACTERIZING RICE PEST AND GENETICS OF RESISTANCE

2.D. Foreign genes as novel sources of resistance to Rice Hoja BlancaVirus and Rhizoctonia solani

L.Calvert (IP-4), E.Tabares (SB2), G.Delgado (IP4), L.F. Fory (SB2, IP4,María Angélica Santana (SB2 1), Nilgun Tumer 2, Z. Lentini (SB2, IP4).

2.D.1. Introduction

The main goal of this project is to provide new source(s) of resistance tocomplement the single breeding resistance source present in most of thecommercial varieties grown in Latin America nowadays. This breeding source isconferred by one or two genes, and does not protect plants at ages younger than25 day-old. The project aims to transform rice with novel gene(s) for RHBVresistance, and to incorporate these genes into Latin American commercialvarieties or into genotypes to be used as parents in breeding. Previous reportsdescribed the generation and selection of Cica 8 transgenic plants carrying thenucleoprotein (RHBV-N) viral gene. The resistance conferred by the N gene ischaracterized by a significant delay in the progression and severity of diseaserespect to inoculated non-transgenic controls. Advanced generations oftransgenic lines with stable RHBV resistance has been selected. In contrast, theCica 8 non-transgenic control is highly susceptible throughout the whole lifecycle, showing severe disease development and most plants die at 60 days afterinoculation. Last year, results also suggested that the resistance conferred by theN transgene towards RHBV disease is expressed independently of the genotypebackground. The transgenic resistance could be used to complement the naturalresistance source to the virus, when crossing selected transgenic lines withdiverse genotypes carrying the breeding resistance gene(s). Results showedthat the non-transgenic F1s control plants were susceptible, whereas thetransgenic F1s were resistant even when inoculated at 10-day-old. These resultssuggested that the protection conferred by the RHBV-N transgene is inheritedand expresses independently of the genotype background, and that thetransgene could be used to complement the natural resistance source. A total of421 selected transgenic lines representing various generations, and F2populations derived from crosses with FB007, Oryzica 1, Iniap 12, and Cica 8 willbe planted in the field on November 2000. These lines will be evaluated forRHBV resistance and agronomic traits following International as well as theColombian environmental biosafety regulations at Palmira experimental station.The approval for field testing by the Colombian Biosafety Committee was issuedon September 2000. Following it is reported the characterization of modeexpression of the transgenic resistance conferred by RHBV-N, and the progressgenerating transgenic rice containing the RHBV non-structural 4 (NS4) gene fromthe RNA 4.

1 IDEA, Caracas, Venezuela.2 Biotechnology Center, Rutgers University

126

Outputs

2.D.4. Characterization of RHBV-N resistance in transgenic plantsL.F. Fory (SB2, IP4), T. Agrono (IP4), C. Ordoñez (IP4), M. Cruz (IP4), M. Duque (SB2,

IP4), J. Silva (IP4), C. Dorado (SB2),Z. Lentini (SB2, IP4), L.Calvert (IP4).

The level of resistance conferred by the RHBV-N viral gene was evaluated byscoring the percentage of leaf area diseased, severity of symptoms, and vigor.Evaluations were conducted once a week starting 5 days after removal ofviruliferous insect vectors, up to 54 days after inoculation. To conduct this study,preliminary 25 transgenic lines represented by 8 plants each were evaluated at15 days after germination. Of these, eight lines were selected showing high vigor.These lines were used to study the effect of the plant age on the level oftransgenic resistance. Two disease pressures were used, intermediate diseasepressure (colony of 65% of virulence) and high disease pressure (colony of 70%of virulence), and each plant was inoculated with four insects per plant.

Line A3-49-60-12-3-3 showed the highest level of resistance throughout thewhole life cycle. Between 74% to 81% of the plants did not show any diseasesymptoms when inoculated either at 14 days or 28 days of age, and only a 22%of the plants showed more than 25% of the leaf area affected when inoculated at14-day-old (table 1). In contrast, Cica 8 control showed between 70% of theplants with severe disease symptoms at 14-day-old (table 1). Line A3-49-60-4-5-8 showed intermediate level of resistance at 14-day-old and 71% of the plantswithout symptoms at 28-day-old (table 2.36). About 70% of the plants of line A3-49-60-19 had less than 25% of leaf area affected at 14-day-old, whereas Cica 8showed 100% of plants highly diseased (table 2.37). In general, the level oftransgenic resistance increased at 28-day-old when yet Cica 8 control is stillhighly susceptible (table 2.36 and 2.37). Sister lines A3-60-12-3-1 (susceptible)y A3-60-12-3-3 (resistant) showed different disease reaction indicating that theresistant phenotype is still segregating at the T4 generation or gene silencingmaybe affecting the expression of the RHBV-N gene in some of the plants. Acomparative Southern blot analysis using methylation sensitive and methylationinsensitive restriction enzymes will be conducted to elucidate if gene silencing isinvolved in the lost of gene expression.

Analysis was conducted to compare the mode of phenotypic expression ofRHBV-N transgenic resistance respect to the standard breeding resistancesource derived from Colombia 1. Fifteen day-old plants from the varieties IR8,Cica 8, Oryzica 1, Oryzica Llanos 5, Colombia 1, Fedearroz 50 and Fedearroz2000 were inoculated in the field using plots of four rows per variety, and acolony of 70% of virulence. All the seven varieties, including the highly resistantvariety Fedearroz 2000, showed disease symptoms. Although Fedearroz 2000,had the lowest percentage of plants with symptoms and a reduced leaf area

127

diseased. When symptoms developed, including on those of highly resistantvarieties (i.e. Fedearroz 2000), symptoms appeared as the typical white leaf stripor white spots throughout the leaf similarly to the highly susceptible varieties IR8and Cica 8. In contrast, when diseased symptoms developed in resistanttransgenic plants they sometimes appeared as necrotic spots resemblinghypersensitive reaction. Often times these symptoms were mainly observed onthe original leaves were the insects fed onto, but new leaves appeared free ofsymptoms giving a recovery phenotype to the plant. This type of resistantphenotype suggest that the mode of action RHBV-N resistance might berestraining the virus replication or mobility throughout the leaf once the plant cellsare infected.

Table 2.36. Disease resistance on T4 transgenic plants inoculated at14 day-old and 28 day-old

Age atLeaf area affected

(% plants)Line inoculation 0 >0-25 >25-100

A3-49-60-12-3-3 14 74 4 22A3-49-60-4-5-8 14 54 0 46Cica 8 (control) 14 22 9 70

A3-49-60-12-3-3 28 81 19 0A3-49-60-4-5-8 28 71 12 16Cica 8 (control) 28 33 0 66

Between 22 to 24 plants were evaluated per line per total of threeReplications

Table 2.37. Disease resistance on T2 transgenic plants inoculated withhigh pressure at 14 day-old and 28 day-old

Age at Leaf area affected (% plants)Line inoculation 0 >0-25 >25-50 >50

A3-49-60-10 14 47 0 0 53A3-49-60-13 14 25 4 12 58A3-49-60-19 14 53 12 18 18A3-49-56-15 14 9 13 17 61A3-49-60-12-3-1 14 7 0 0 93A3-49-101-18-19-2 14 15 0 8 77Cica 8 14 0 0 0 100

A3-49-60-13 28 44 26 17 13A3-49-56-15 28 29 21 17 33Cica 8 28 0 0 0 100

Between 16 to 24 plants were evaluated per line per total of two replications

128

2.D.5. Generation of transgenic rice containing the RHBV non-structural 4(NS4) gene from the RNA 4

L.Calvert (IP-4), E.Taberes (SB2), G.Delgado (IP4), L.F. Fory (SB2, IP4), María Angélica Santana (SB2 3), Nilgun Tumer 4, Z. Lentini (SB2, IP4).

Earlier studied conducted by Lee Calvert and coworkers at CIAT using specificantiserum had indicated that the major NS4 protein is expressed in rice plants,but not in any of the instars or adult plant-hoppers (Tagosodes orizicolus). Therealso appears to be a control mechanism that allows expression in the plant butnot in the planthopper, since the complete RHBV genome could be isolated fromviruliferous vectors. In contrast the N protein is expressed in both the plant andthe insect vector. It is inferred from the differential expression of these proteinsthat the major NS4 protein may has a function that is needed in the plant but notin the plant-hopper. The differential plant-insect NS4 expression, and thesimilarity of NS4 sequence with well characterized helper proteins described forother insect transmitted viruses, suggest that NS4 might be involved in the RHBVtransmission from the plant to the plant-hopper, or in the virus movement fromcell to cell. The strategy for the expression of the RNA4 in transgenic rice is todetermine the function of the major NS4 protein and study the potential for anovel and different method of producing viral resistant plants.

The RHBV NS4 gene in sense and anti-sense orientations driven by the 35SCaMV promoter were placed into the plasmid pCAMBIA 1301 carrying the gus-intron and hygromycin resistance gene (table 2.38). The NS4 gene in bothdirections driven by the unbiquitin promoter was also cloned into vectors carryingthe hygromycin-cat 1 intron gene from Peter Waterhouse’s laboratory at CSIRO,Australia (table 2.38).

The indica rice varieties Cica 8 (control for transformation efficiency), Palmar,and Cimarrón are being used as targets. Palmar (high grain/ milling quality) andCimarrón (high yielding variety) are commercial varieties from Venezuela highlysusceptible to RHBV. Part of the project is funded by the Centro TecnológicoPolar, Venezuela. A total of 24 transgenic plants carrying the NS4 senseorientation, and 25 plants carrying NS4 anti-sense orientation driven by the 35SCaMV promoter were produced from independent events by Agrobacteriummediated transformation using the Agl1 strain. Southern analyses using Bam HIor Eco RI which excise the complete NS4 gene in sense or antisense orientation,or using Sal I which does not cut the gene cassette within the right and leftborders were used. Results indicate that 100% of the plants had integratedsingle non-rearranged copy of the gene (Figure 1). Northern and Westernanalyses will be conducted and RHBV resistance evaluations will proceed nextseason.


129

Table 2.38. Description of RHBV-NS4 constructs generated at CIAT

Name Gene Promoter Vector/orientation Other genes

pIC001 NS4 35S CaMV PC1300/sense hpt, Kan

pIC003 NS4 35S CaMV PC1300/asense hpt, Kan

pIC002 NS4 35S CaMV PC1301/sense hpt, Kan,

GUS-intron

pIC004 NS4 35S CaMV PC1301/asense hpt, Kan,

GUS-intron

pIC005 NS4 Ubiquitin NT168/sense --

pIC006 NS4 Ubiquitin NT168/asense --

pIC007 NS4 Ubiquitin PWBVec8/sense hpt-cat intron

pIC008 NS4 Ubiquitin PWBVec8/asense hpt-cat intron

pIC009 NS4 35S CaMV PWBVec8/sense hpt-cat intron

pIC010 NS4 35S CaMV PWBVec8/asense hpt-cat intron

2.D.6. Fungal genes as novel sources of resistance to Rice Hoja BlancaVirus and Rhizoctonia solonia.

E.Tabares (SB2), G.Delgado (IP4), L.F. Fory (SB2, IP4,María Angélica Santana (SB2 5), Nilgun Tumer 6, Z. Lentini (SB2, IP4).

The fungal complex composed Rhizoctonia solani (sheath blight),Helmithosporium, Rhincosporium, and Sarocladium is already causing importantrice yield losses in the Southern cone of South America and increasing spreadshad been reported in Colombia, Mexico and Venezuela. All rice varieties aresusceptible and there are not known sources of stable genetic resistance forthese diseases in rice. In the case of sheath blight, IRRI had placed a majoreffort in developing biological control strategies for this disease without successeither. At present, the control of this complex mainly depends on use of


130

fungicides (Dr. Fernando Correa, CIAT Rice Pathologist, Cali, Colombia,personal communication). Recently, FLAR suggested CIAT (Dr. Peter Jennings,personal communication), to develop molecular strategies for incorporatingresistance to this fungal complex. However, very little is known about theinteraction between the rice and these pathogens in order to direct specificresistance strategies for each of these fungi. Of the four, the plant-pathogeninteraction with Rhizoctonia solani is the better known.

Work conducted by another principal investigator of this project (Dr. NilgunTumer, Biotechnology Center at Rutgers University, USA) showed that apokeweed antiviral protein (PAP), a 29-kDa protein isolated from Phytolaccaamericana (a weed naturally found from USA to Argentina), has a ribosome-inactivating ability. Mutated versions of PAP gene has potent antifungal activity(Zoubenko et al., 1997). Homozygous progeny of transgenic tobacco plantsexpressing these PAP genes displayed resistance to the fungal pathogenRhizoctonia solani. Transgenic PAP potato showed protection againstPhytophtora infestans, and transgenic PAP turfgrass are resistant to variousfungal pathogens. These results suggest the possibility of designing molecularstrategies for incorporating fungal resistance by introgression of mutant PAPgene(s) in transgenic rice plants. Here we report the progress made during thefirst year of this project.

Outputs

2.D.7. Production of rice plants carrying various versions of PAP gene

Indica rice varieties Cica 8 (control for transformation efficiency), Palmar,Cimarrón, and Fundarroz PN1 are used as targets. Palmar and Cimarrón showshigh and moderate tolerance to sheath blight, whereas Fundarroz PN1 and Cica8 are highly susceptible to sheath blight. The strategy includes to evaluate themode of action of PAP in highly susceptible as well as in tolerant sheath blightgenotypes, to determine if PAP could increase the level of protection. Thisproject is supported by the Centro Tecnológico Polar, Venezuela. To generatenew point mutations in the PAP gene, a rapid change site directed mutagenesiskit from Stratagene was used. Gene constructs carrying various mutant versionsof the PAP gene, were placed in vectors driven by 35S CaMV promoter or maizeubiquitin promoter, and using hygromycin resistance as gene selection.

Eight new PAP mutations were generated directed to change the aminoacidscomposition in the PAP protein (table 2.39). These new mutated genes wereplaced into yeast vectors, and transformed into yeast to check for no toxicity. Thenon-toxic mutated genes are being transformed into tobacco first to check thegene expression and toxicity before using them for rice transformation.

Two mutated versions of PAP (I deleted and II) already tested for no toxicity inturfgrass (another monocot species) are being used as the first approach to

131

transform rice (table 2.40). These genes driven by the ubiquitin promoter wereplaced in the plasmid vectors pWBVec8, pWB10a, and pBGXiHGFP kindlysupplied by Dr. Peter Waterhouse (CSIRO, Australia). These plasmids had beenused successfully by Waterhouse to transform rice via Agrobacterium. Theycontain a hpt gene with a CAT-1 intron for increased expression of hygromycinresistance and selection in rice, a gus-intron-gene, or a gfp (green fishfluorescent) gene, respectively, to aid the recovery of transgenic plants.

A total of 35 independent transgenic events carrying the PAPI deletion mutantgene, and 50 independent transgenic events carrying the PAPII gene had beengenerated up to now. A first set of plant tissue was sent to Rutgers this summerfor analysis and plants with PAP gene expression were identified based onWestern analysis (Figure 2.8). PAP expressing plants will be evaluated forsheath blight resistance under greenhouse conditions, while detailed molecularanalyses are being conducted to determine the number of gene copy andpatterns of integration into the rice genome.

132

Table 2.39. New mutations generated in PAP gene.

Name in PlantVector

Name in YeastVector

Mutation AA Change

NT296 NT299 PAPI Del I4M

NT298 NT300 PAPI Del T18M

NT317 NT311 PAPI Del I13M

NT319 NT312 PAPI Del V8M

NT NT PAPI Del Y16M

NT NT PAPIPoint

Y16A

NT NT PAPIPoint

Y16S

NT NT PAPIPoint

Y16Phe

Table 2.40. Description of PAP constructs for Plant Transformationgenerated

Name GEN Promoter Vector/orientation Other genes

NT178 PAPId Ubiquitin NT168 --

NT301 PAPId Ubiquitin NT294 GFP, hpt-catintron

NT303 PAPId UbiquitinPWBVec10a

GUS, hpt-catintron

NT306 PAPId Ubiquitin PWBVec8 hpt-cat

RT126 PAPII Ubiquitin NT168 ---

NT302 PAPII Ubiquitin NT294 GFP, hpt-catintron

NT304 PAPII Ubiquitin PWBVec10a GUS, hpt-catintron

NT305 PAPII Ubiquitin PWBVec8 hpt-cat

133

2.D.8. Training Activities and Conference OrganizationA.Mora (IP4),G.Delgado (IP4), T. Agrono (IP4), C. Ordoñez (IP4), E.Tabares (SB2),

L.Fory (IP4, SB2), Z. Lentini (SB2, IP4).

2.D.8.1. International Courses

• Advanced course on integrated application of plant breeding and moleculartechniques for rice bredding. September 2000. CIAT, Cali, Colombia.

• Coordination of International workshop. Development of Insect and FungalResistant Rice: Introgression of genetic resistance to pests and diseasesdependant on chemical control”. April 23-28. Porto Alegre, Brazil.

2.D.8.2. National Courses

• Coordination of Workshop on Agriculture Biosafety for the ColombianNational Biosafety Committee. April 13-15, 2000. CIAT, Cali, Colombia

2.D.8.3. International Training at CIAT• Ing. José Antonio García. Universidad de Costa Rica.• Ing. Mervin Vargas. Oficina de Semillas. Costa Rica.

PAPI 50 49 48 47 46 45 44 43 42 41 40 39

FIGURE 2.8. WESTERN BLOT ANALYSIS OF TRANSGENIC PALMAR CARRYINGDELETION MUTANT GENE. ANALYSIS SHOWED PLANTS WITH VARIOUSOF PAP

134

• Dr. Miguel Muñoz Fonseca. Universidad de Costa Rica• Luis Orlando Duque. Agronomy Engineering Thesis completion 2000.

Universidad Nacional. Sede Palmira. Outstanding Recognition.

Publications

Chapters in Books• Lentini, Z. 2000. Biotecnología en el fitomejoramiento del maíz. 207-243. In:

H. Fontana and C. González (Eds.). Maíz en Venezuela. Fundación Polar.Caracas, Venezuela. 529 p.

In Refereed Journals• Lentini Z., Lozano I, Tabares E., Fory L., Domínguez J., Cuervo M., Calvert L.

2000. Expression and inheritance of hypersensitive resistance to rice hojablanca virus mediated by the viral nucleocapsid protein gene in transgenicrice. Theoretical and Applied Genetics (submitted June 2000, accepted withrevision September 2000)

In Non-Refereed Publications• Newspaper report by Laura Tangley. Engineering the harvest. Biotech could

help fight hunger in the world’s poorest nations- but it will?. U.S. News &World Report. March 13, 2000. Reports work on transgenic rice at CIAT.

• Information Systems for Biotechnology ( ISB) News Report. May, 2000 byTimothy Pratt. Colombia Biosafety Council takes on training. Report onbiosafety workshop offered to the Colombian Biosafety Committee.

Presentation in Workshops, Conferences, Meetings and Posters• April 25, 2000. Biotechnology Applied to Rice Germplasm Development. XI

Fenarroz (Feria Nacional de Arroz y Muestra Comercial e Industrial). InvitedSpeaker.

• April 25, 2000. Transgenic Crops: Environmental and Food Biosafety. XIFenarroz (Feria Nacional de Arroz y Muestra Comercial e Industrial). InvitedSpeaker.

• April 26, 2000. Biotechnology for Broadening the Genetic Base of Rice inLatin America. Workshop. Development of Insect and Fungal Resistant Rice:Introgression of genetic resistance to pests and diseases dependant onchemical control”. April 23-28. Porto Alegre, Brazil.

• September 1, 2000. Forum on “Transgenic plants an alternative for plantpathogen management: Risks and Benefits”. XXI Ascolfi Congress. August20- September 1, 2000. CIAT Cali, Colombia.

• September 21, 2000. Industrial uses of transgenic plant: Current andPerspectives. Forum on Food Biosafety and Biotechnology Development:Risk and Opportunities. ILSI-Nor Andino. Universidad Javeriana. Bogotá,Colombia.

• October 16, 2000. From germplasm banks to farmers fields: Role of CIAT

135

biotechnology in research and training in Latin America. Invited Speaker atpioneer Hi-Bred Seeds. Johnston, Iowa. USA.

136


2.E. Characterization of Entorchamiento: A Complex of Polymixa graminisand Rice Stripe Necrotic Virus

L. Reyes, G. Prado, R. Sedano, F. Correa, F. Morales and L. Calvert

2.E.1. Introduction

During this year, priority research for entorchamiento (crinkling) disease, causedby rice stripe necrosis virus (RSNV) and vectored by Polymyxa graminis, hasbeen the development of a screening methodology that gives consistent resultsand is able to evaluate sufficient lines that can be incorporated into a breedingprogram. In addition, we are continuing to monitor the spread of the disease inColombia as well as other countries. To minimize the spread of the disease,there is a high demand for a seed treatment and we report on some preliminarywork on chemical control of the fungal vector.

2.E.2. The development of methodologies to screen rice for resistance toentorchamiento

Entorchamiento (crinkling) disease caused by a complex of rice stripe necrosisvirus (RSNV), and its vector Polymyxa graminis. During this year, the priorityresearch has been the development of a screening methodology that givesconsistent results and is able to efficiently evaluate lines.

One previous method used sand to grow the plants and depended on inoculationwith pulverized roots infected with P. graminis. At regular intervals the waterlevels were raised and lowered. The plants tended to suffer under the stressfulconditions of a hydroponic system and the levels of RSNV incidence weregenerally 50-70%. Another method depended on using soils contaminated. Thismethod was often superior to the hydroponic method but the results wereinconsistent, and It took up to two months for the development of symptomsusing either method.

The current methodology is a combination previous methods and was developedby side by comparisons and statistically analysis of the results. The first majorchange was to plant the rice in small trays using contaminated soils. The soil iskept humid and allows the exposure of the germinating roots to the zoospores ofthe fungus. Twelve days after planting, the rice is transplanted to pots (10 plants/pot) that contain a mixture of contaminated soil (50%) and sand (50%). Thetransplanting produces small wounds on the roots facilitating the entry of thezoospores. The contaminated soil was inoculated with pulverized rootscontaining cistosoros of the vector. The pots are maintained in a larger plexi-glass tray and infected plants are grown with the test plants to maintain an active

137

source of zoospores. The results are read at 30-40 days after planting or 25 daysafter the rice is transplanted.

The results were that the method using contaminated soils is superior to usingsterilized sand or soil (table 2.41). The method was superior in the percentage ofincidence and time needed for the maximum expression of the disease. Therewas also less variance between the replications using the new methodology.

Table 2.41: Comparison of two methods to evaluate rice germplasm forresistance to entorchamiento

Variety

Sterile soilsinoculated withpulverized rootcontaining P.graminis/RSNV

Days tillmaximumincidence

Contaminatedsoils inoculatedwith pulverizedroot containingP. graminis/RSNV

Days tillmaximumincidence

T-Studenttest

entorchamiento +% +/- std err

entorchamiento +% +/- std err

Orizyca Llanos5

66 +/- 22 52 95 +/- 5 31 p < 0.001

Orizyca 3 74 +/- 18 52 91 +/- 8 29 p < 0.01Fedearroz2000

70 +/- 25 60 75 +/- 11 39 p < 0.45

2.E.2.1. Screening germplasm in field conditions

At this time, there is no effective mechanical inoculation for entorchamiento.Therefore a field that was severely affected by entorchamiento (more than 90%incidence) during the previous crop of rice was selected to make a fieldevaluation of 233 rice lines. The materials included commercial varieties,promising rice lines and the VIOFLAR collection. The field was prepared usingcultivation practices that are favorable to the development of entorchamiento.The rice lines were grown in rows and there were three replications of each line.The rice lines were planted in a randomized block design. The evaluationconsisted of counting all the plants in a row and the number of plants showingsymptoms of entorchamiento. The field was evaluated at approximately 40 daysafter planting

The incidence of the disease was generally fairly low. Only 10 lines had morethan 20% of the plants infected. The three most infected varieties were O. llanos5 with 46.3% incidence, FONIAP 1 with 27.9% incidence and Oryzica 3 with23.9% incidence. Although a further analysis is needed, it appears the fieldresults are only partly consistent with the greenhouse evaluation. For example,Fedearroz 50 had 10% incidence in the field and 75% incidence in thegreenhouse. FONIAP 2000 had an incidence of entorchamiento of only 1.7% inthe field but 70% in greenhouse conditions.

138

Since entorchamiento is a complex of both P. graminis and RSNV, it may bedifficult to develop a field screening technique that is reliable. This experimentclearly demonstrates that the level of incidence of entorchamiento is highlyvariable in each cycle of rice. To make an effective screening sufficient diseasepressure is essential. The challenge will be to develop tanks or field plots inwhich high disease pressure can be maintained during many cycles of rice.

2.E.2.2. Screening germplasm for resistance to RSNV

After developing a methodology to screen for resistance, a larger sample of ricelines were tested with either three or four replications, each containing 10 plants(table 2.42). These included 13 commercial varieties, 55 rice lines out of which48 are part of VIOFLAR germplasm bank, and one source of O. glaberrima.Susceptible controls were O. Llanos, 5 with an average incidence of 88% andOrizica 3, with an average of 86% of the plants infected during 11 replications.Variety Colombia 1, which is reported to be resistant had an incidence of 34%.Varieties Fedearroz 50 and Fedearroz 2000 were both highly susceptible toRSNV. About 50% of varieties had an incidence that was equal or better thanColombia 1, with 18% of lines with less than 20% infection. The only materialsthat are apparently immune to RSNV were, as previously reported, O.glaberrima. The recently released varieties Colombia XXI and Coprosem 1 haveexhibited different levels of resistance/susceptibility in different trials includingdifferent sources of inoculum. These results could suggest the existence of racesin the pathogen complex, which deserves more research in the near future todetermine the real reaction of these cultivars to the disease.

Most of the commercial Colombian rice cultivars exhibit different levels ofsusceptibility to the virus under field conditions but those levels have not beenwell characterized under controlled conditions. Aiming at identifying resistancesources to entorchamiento we screened in a separated experiment in 2000 atotal of 202 rice lines including advanced lines of the breeding program,Colombian commercial cultivars, and wild rice species (Table 2.43). Inoculumwas based on infested soil collected from farmers’ fields during epidemicdevelopment of the disease in 1999. All symptoms typical of entorchamientowere observed including yellowing or chlorosis, stunting, crinkling, and deadplants (Table 2.43). No resistance sources were identified among the advancedlines and commercial Colombian cultivars.

Rice lines in this study were not breed or selected because of their reaction toRSNV. Therefore, number of rice lines with a degree of resistance to RSNVimplies that this type of screening can be made part of a breeding program forrice, without being too much of a bottleneck for germplasm. If some of the mostresistant lines are used as a parent, even more or the progeny, should beresistant to entorchamiento.

139

Table 2.42. Distribution of percentage RSNV infected plants in 69 ricelines

Incidence of diseased plants0-20% 21-40% 41-60% 61-80% 81-100% Total

Rice lines evaluated 12 24 9 20 4 69

% of lines showing resistance 17.4% 34.8% 13.0% 29.0% 5.8% 100%

Table 2.43. Incidence Range (%) of Rice Stripe Necrosis Virus(Entorchamiento) Symptoms in Greenhouse Evaluations of 202 Rice Lines

Rice Lines1 LinesNo.

Stunting%

Crinkling%

Dead Plants%

Advanced lines 178 45-100 5-83 0-81

Colombian commercialcultivars 13 53-93 18-58 0-50

Wild species (3)accessions 11 0-51 0-3 0-30

1A maximum of 50 plants evaluated per line in 10 replications.

2.E.3. Testing Wild Species for Resistance to Entorchamiento

High level of resistance to entorchamiento was identified only among the wild ricespecies where 11 accessions of Oryza glaberrima were immune to the virus(Table 2.44). These lines did not exhibit any of the symptoms typical ofentorchamiento. Oryza barthii and Oryza rufipogum were more resistant than theadvanced lines or commercial cultivars, however only one accession from eachspecie was tested (Table 8). O. rufipogum exhibited high levels of stunting. SomeIRAT advanced lines also showed good levels of resistance and are being testedagain for confirmation of their reaction. Interspecific populations developedbetween O. glaberrima and O. sativa through backcrossing will be tested underthis greenhouse methodology for the identification of resistance sources. Thesepopulations will also be used for the identification of molecular markersassociated with the resistance to the virus.

140

Table 2.44. Incidence (%) of Rice Stripe Necrosis Virus (Entorchamiento) inGreenhouse Evaluation of Wild Rice Species

Wild Species PlantsNo.

Stunting%

Crinkling%

Dead Plants%

Oryza rufipogon 33 52 3 30

Oryza barthii 42 0 2 2

Oryza glaberrimaAccession 1 52 0 0 05486 TOG 8/98 48 0 0 0CG-20 8/98 49 0 0 06405 TOG 8/98 47 0 0 0CG-14 8/98 49 0 0 0IG-14 8/98 46 0 0 05810 TIG 8/98 50 0 0 05980 TOG 9/98 43 0 0 06331 TOG 9/98 51 0 0 0

2.E.4. Chemical control of Entorchamiento

Due to the high levels of entorchamiento observed in many commercial rice fieldsin different areas of Colombia, rice farmers are using different chemicalsincluding fungicides and insecticides for the control of the disease. However, theefficiency of many of these chemicals in the control of entorchamiento has notbeen demonstrated. A new systemic fungicide developed in Japan againstseveral soil pathogens was tested for the control of entorchamiento in thegreenhouse. This fungicide is also known as promotor of plant development androot growth and is recommended for soil or seed treatment. Healthy seeds of thesusceptible rice cultivar Oryzica 3 were planted in infected soil under differentseed or soil treatments with this fungicide (Table 5). The most efficient treatmentfor the control of entorchamiento was the 1X and 5X doses of the fungicide assoil treatment (Table 2.45). The 10X-soil treatment had low levels of crinkling andstunting, however the level or incidence of yellowing symptoms was higher thanthe two reduced doses. We will determine in further studies if the yellowingobserved in these treatments is a symptom of the disease or if it is a partialphytotoxic effect of the fungicide. The seed treatment at least at the doses usedwas not effective in the control of the disease. Further experiments are beingconducted in the greenhouse as well as in the field to corroborate our resultsbefore the recommendation of the product.

141

Table 2.45. Seed and Soil Fungicide Treatment for the Control of Polymyxagraminis Vector of Rice Stripe Necrosis Virus (Entorchamiento)

Treatment1PlantsNo.2

HealthyPlants

%Stunting

%Crinkling

%

DeadPlants

%

T1 1X Seedtreatment 107 18 38 57 23

T2 5X SeedTreatment 106 8 55 65 15

T3 10X SeedTreatment 109 11 46 52 11

T4 1X SoilTreatment 118 77 9 11 3

T5 5X SoilTreatment 122 78 7 7 0

T6 10X Soiltreatment 111 59 8 6 0

T7 5X Seed + 5XSoil treatment 116 72 6 6 0

T8 Untreated 107 9 38 53 10

1Each treatment consisted of 10 replications.2Fungicide treatments – IX, 5X, and 10X of active ingredient under experimentation.Oryzica 3 was used in all tests.

2.E.5. Monitoring the range and incidence of rice stripe necrotic virus

The range of rice disease ‘entorchamiento’ is quickly spreading. It is now inPanama and Costa Rica. Vector of the virus is Polymyxa graminis and thesporangia are transmitted on the seed coat. Successful new varieties, such asFedearroz 50, are being imported into other countries and this is causing a rapidincrease in the range of the causal agent rice stripe necrotic virus (RSNV).

In Colombia the disease took 7-10 years to become widespread. It is now in allmajor rice growing regions and must be considered endemic. If Colombia wantsto maintain a seed export business, certification that fields are free ofentorchamiento is needed.

142

During the first semester of year 2000, the Rice Stripe Necrosis Virus"Entorchamiento" was reported, for the first time, in the rice-growing area of theCaribe Seco, at the municipalities of Pivijay and Valledupar, departments ofMagdalena and Cesar, respectively. The RSNV was observed in differentcommercial varieties and rice lines which exhibit susceptible reactions(Fedearroz 50, Oryzica 1, CNAX5013-13-2-2-4-M , CT11275-3-F4-8P-2,CT11408-6-F4-1P-3, CT11275-4-M-1-M, CT11408-6-F4-14P-2, LV223-1-1-3-8-3-M, FSR456-M-1-2-2).Last year in the Cauca Valley, there was a high incidence of entorchamiento,however, farms that are being monitored indicate that the incidence is muchlower this year. Rice growers have made modification in the preparation of theland and that appears to be reducing outbreaks of entorchamiento.

2.E.6. Molecular characterization of RSNV

RSNV is a member of the Benyvirus group. The type member of this group isbeet necrotic yellow vein virus (BNYVV). Another member of this group is beetsoil-borne mosaic virus (BSBMV). A 2100 base pair cDNA clone representing theRNA 1 of RSNV was sequenced and compared for nucleic acid and amino-acidhomology with these benyviruses. The degree of homology varied greatly withthe region near the 5’ terminus of the RSNV clone having the lowest degree ofhomology and the 3’ terminus having the highest degree of homology. Theresults of the comparison are shown in table 2.46. The complete region for whichthere is sequence data has fairly low homology with BNYVV and BSBMV with theexception of the last 500 bases. This is a region that is more highly conservedbetween the three viruses. BNYVV and BSBMV are more closely related to eachother than to RSNV. These results confirm the relationship of RSNV as amember of the benyviruses, but one that is only distantly related to the otherknown viruses in the group.

Table 2.46. Amino acid homology of a portion of the RNA 1 of threemembers of the benyvirus group

Virus BNYVV2100

base region

BSBMV2100

base region

BNBVV500

base region

BSBMV500

base region

RSNV RNA 40.3%1 40.2%RSNV AA 76.4% 77.7%BNYVV RNA 73.1%BNYVV AA 92.3%1Percentage of identity of nucleic acids or amino acids.

143

2.E.7. Training

Different meetings, courses, and other training events on IPM, particularlyregarding T. orizicolus and the hoja blanca virus, have been held with theparticipation of farmers and agronomists.

2.E.7.1. Conferences

Date Locality Lecturer TopicNovember 1999 Cúcuta L. Reyes RHBV-Tagosodes orizicolus and

RSNVDecember 1999 Geneva L. Reyes RHBV-Tagosodes orizicolus and

RSNVFebruary 2000 Jamundí L. Reyes RHBV-Tagosodes orizicolus and

RSNVMarch 2000 Ibagué M. Triana Tagosodes orizicolus /RHBVMay 2000 CIAT M. Triana RHBVSeptember 2000 CIAT M. Triana

Tagosodes orizicolus /RHBVSeptember 2000 CIAT1 M. Triana Breeding for resistance to insects and

evaluation of RHBVSeptember 2000 CIAT1

L. ReyesIntegrated management of Tagosodesorizicolus and selection of advancedlines for resistance to RHBV

September 2000 CIAT1 R. Meneses IPM1. Course on Application of Conventional and Molecular Methods to Rice Improvement, held at CIAT, 24 September-6

October 2000.

147

OUTPUT 3. ENHANCING REGIONAL RICE RESEARCH CAPACITIES ANDPRIORITIZING NEEDS WITH EMPHASIS ON THE SMALL FARMERS

3.A. FLAR and Economics of Rice Production Systems

3.A.1. Manage FLARL.R.Sanint, Executive Director FLAR and Rice Project Economist, CIAT

Marco A. Oliveira, Deputy Director for Administration, FLAR, Southern Cone.

3.A.1.1. Organize and direct meetings of the administrative and technicalcommittees

Milestones

One meeting of the Administrative Committee: Oct. 1999 (Montevideo). Nov.2000, CIAT.Two meetings of the Technical Committee.

- Tropical region: Santa Rosa (Colombia), August 2000.- Temperate region: Cachoeirinha (Brazil), June 2000.

Administrative Committee. the eighth meeting had its deliberations in Montevideo,Uruguay, during the month of October, 1999. This year, the ninth meeting will takeplace at CIAT, Nov. 30 to Dec. 1, 2000.

Technical Committees

FLAR separated its technical meetings into two Subcommittees, one for each ofthe main ecoregions: the tropical and the temperate.

The meeting of the temperate region took place at the Experiment Station ofCahoeirinha, in Brazil, during June 22-23. The committee agreed on a new flow ofgermplasm. Its development for the temperate region has to emphasize coldtolerance. Progenitors from Uruguay, Chile, Rio Grande do Sul and USA will becharacterized at CIAT under controlled settings. Triple crosses will be done inColombia and the material will be processed by anther culture, where they will beevaluated for blast and quality. The R3 lines will be evaluated in Uruguay (some10,000 lines). The resulting R4 will conform the VIOFLAR that will be distributed toall members.

The tropical committee met at the Santa Rosa Station in Colombia on August 25.All country members from the tropical region were present. The group stressedthe need to revise the strategy for blast resistance. Emphasis on other diseases isalso a must. FLAR should monitor developments in hybrid rice.

3.A.1.2 Seek new members and expand participation in current membercountries.

148

During 2000, FLAR maintained contact with key rice people in nonmembercountries such as Peru, Ecuador, Mexico and the Dominican Republic.Conversations are most advanced in Peru and Ecuador, but their entrance maytake one more year as internal organization is still weak.

3.A.1.3. Personnel management, supervision, recruiting.

Departures

Dr. Takazi Ishiy, main breeder for the Southern Cone, left FLAR on August 31,2000, as the contract between EPAGRI and IRGA was cancelled due to internaldisagreements.

Liz D. Arango, FLAR's secretary, left on October after five years of service.

Arrivals

Dr. Carlos Bruzzone, former CIAT's Rice Project Breeder, is now a part time abreeder with FLAR.

Maribel Cruz, research assistant, will work on characterization of the germplasmbank and identification of promising progenitors for cold tolerance.

María Victoria Ballesteros is the new secretary.

3.A.2. Breeding at FLAR

3.A.2.1 Plan, organize, direct breeding and selection activitiesLuis E. Berrío, Associate plant breeder, FLAR

Peter Jennings, main breeder, FLAR (consultant)Carlos Bruzzone, breeder, tropical region, FLAR (consultant)

Takazi Ishiy, breeder, temperate region, FLARMarco A. Oliveira, Deputy Director Administration, FLAR, Southern Cone.

Collaborators: twelve FLAR member countries, CIAT, IRRI and CIRAD.

Progress in FLAR Breeding activities this year:

INGER-LAC

Access to elite lines is fundamental in any research program. From its creation,FLAR associates gave high priority to INGER-LAC continuity. We have twoclasses of nurseries to be used for exchanges: (i) the VIOAL, formed by publicmaterials, and (ii) VIOFLAR, configured with own materials, whose use isrestricted to the associates. VIOAL depends basically on the germplasmoriginating from INGER-Global (from IRRI), on the National Programs and on linesthat CIAT Rice Project develops through wild germplasm, introgression ofagronomic traits from new plant type and the development of populations through

149

recurrent selection. VIOFLAR is formed by advanced materials from our partnersthat are furnished for research, other material exchanged through direct contactswith multiple rice institutions of the whole world, in addition to lines generated byown improvement activities.

VIOAL 1999, dispatched to countries of the region, has a total of 130 lines, out ofwhich 62 (47,7%) are from National programs of the region; 48 (36,9%) originatefrom INGER-Global Nurseries and the remaining 20 lines (15,4%) were developedby CIAT. Table 3.1 indicates number of dispatched sets and data received, aswell as percentage of lines selection in each localition. Information was onlyreported in 39% of sets. From there, results of a total of 77 lines (59,2%) selectedfor yield trials during the year 2000 were extracted. In 4 of 5 localities thatreported selection, the following lines were outstanding for their good behavior:RCN-B-94-19 and RCN-B-93-83 from Surinam's national program; CT13503-M-13-1-M-5-3P, CT9509-17-3-1-1-M-1-3P-1 (lines used primarily as progenitors intropical crossings), and CT10310-15-3-2P-4-3 (released this year as Fundarroz-PN1 in Venezuela).

VIOAL- Acid Soils 1999, was formed with 28 lines originating from CIAT RiceProject (8 lines), CNPAF/ EMBRAPA (5 lines) and CIRAD-France (3 lines and 12populations of recurrent selection). Even though it was dispatched to 5 countries,only Argentina reported results from INTA-AGUILARES, where 6 lines wereselected for later trials.

VIOAL 2000, was formed with a total of 64 lines as follows: 40 lines originating inINGER-Global Nurseries1999, 3 lines supplied by USA, 4 by Ecuador, 6 by Peruand 11 lines developed by CIAT (inter-specific crossings). A drastic decrease ofmaterial supplied by national programs to the network is observed. Table 3.2indicates number of VIOAL sets dispatched this year by request from severalregional countries.

VIOAL 2001: For this nursery we have 151 lines supplied by the nationalThailand program (Dr. S. Sarkarung, IRRI), 11 of Peru and lines supplied by theCIAT's Rice Project and by CIRAD (Drs. Martínez, Lentini, Valés and Chatel).

This year, it was not possible to introduce to Colombia the INGER-GlobalNurseries that are distributed from IRRI (and that have been introduced for over20 years), due to quarantine requirements of Sanidad Vegetal of ICA, to whichthe Government of Philippines/ IRRI could not answer. Instead, the material wasimported into Venezuela through Fundarroz- (a FLAR partner) and we were ableto introduce 4 types of nurseries, formed by some 380 advanced lines. Thenurseries were planted in June in Barinas (diseases) and in Portuguesa, for theircharacterization. Selection and harvest of most promissing materials by the endof October should be accomplished. Seed from Venezuela will be shipped toFLAR in Palmira to complete their characterization in Colombia and include themin the VIOAL and/or use them as progenitors.

150

Exchange with other sources

In accordance to needs of associates of temperate zone, exchange of lines withresistance to cold and of top quality materials has increased. This year, materialwas received as follows:

Uruguay = 29 materials (6 varieties plus advanced lines)IRGA = 24 materials (including 13 varieties)Chile = 30 materialsUSA = 17 material (3 varieties, 3 lines from Argentina and 11 CT lines).

VIOFLAR nurseries were dispatched to all associates and they were planted in agreat number of localities. Details of these nurseries are included in FLAR’simprovement activities report, below.

Varietal Improvement at FLAR

Santa Rosa: FLAR continues to use the Experimental Station of Santa Rosa, inVillavicencio, Colombia, as the principal site for improvement under favoredupland conditions. High natural incidence of blast ensures an appropriateselection and characterization to such disease. To assure uniform blast pressure,susceptible cultivars are sown in row spreaders 20-25 days before sowing theexperimental material, perpendicularly to that material. The breeding lines aremixed with a very susceptible cultivar called "Fanny".

This last step not done this year, since different sources Fanny were in theprocess of being monitored by the Pathology rice team for their reaction to the 6established lineages found in Santa Rosa. According to results and suggestionsby this team, we have begun, in this semester, the multiplication of a Fanny seedsource, originating from IRRI, and will have available seed for next year. Eventhough Fanny was not intercropped with advanced materials, there was excellentleaf blast pressure blast to evaluate the material sown by FLAR this year (tables3.3 and 3.4).

In the year 2000, FLAR sowed a total of 5,991 lines in Santa Rosa, represented in784 different crossings. This material includes Progenitors (BCF), VIOFLAR,2000-Tropic, F4 Lines, F2 populations and lines derived from anther culture (table3.3).

The F3 seed selected from F2-tropical populations were first evaluated in Palmirafor Hoja Blanca Virus resistance (HB), White Belly (WB) and GelatinizationTemperature (TG). F3 lines selected were multiplied in Palmira from October toFebruary (to obtain the F4) and sown in Montería (November-February) forevaluations mainly of tillering (vigor), lodging and White Belly.

Resulting F3 seed of selected F2-Temperate populations will be evaluated inPalmira to WB, TG, as well as grain length and amilose content. The rest of the

151

seed will be dispatched to IRGA in Brazil, together with R3seed (CA) that wasselected. In Palmira, F2Temperate seed of (586 populations) and 100 "R” linesharvested this year in June and July and will be dispatched. All these materialswill be available for their selection by member countries from the Southern Conein March of next year at the Cachoerinha-IRGA Station, Brazil.

F5 masal seed resulting from selection on F4-tropical, will be evaluated again toWB, TG (Santa Rosa seed), HB, Tagosodes, and milling (including evaluations oflag simulation in the harvest) with seed produced in Palmira. The best resultinglines of these evaluations will be included in VIOFLAR-Tropic 2001 that will bedispatched to the associate countries.

Table 3.4 presents a summary of a total of 2,430 introductions (351 crossings)sown during this year, which include: VIOAL 2000, national Peruvian Programs,Thailand and Venezuela, material of CIAT Project and lines selected in thecountries in VIOALES 1997 and 1999. Selected lines will be characterized inPalmira to: WB, TG, Amilose, HB, Tag and milling, and the best ones will bedispatched in VIOAL 2001 through INGER-LAC to all the countries that request it.

In Palmira the same materials were planted in the May-September/ 2000 cycle,with exception of F2 populations, to obtain clean seed for future evaluations and/ordispatches.

Germplasm Bank: We have consolidated only one Bank, called BCF (CIAT-FLARGermplasm Bank). At present, we have 1,742 entries. Of these, we have thefollowing working collections: 250 commercial varieties, 517 irrigation entries and227 upland entries. Most of the BCF entries are being characterized during thisyear in Calabozo (Venezuela) and in Montería, Saldaña and Villavicencio(Colombia).

This year a new cold room was adapted with ideal conditions (Temperature =14ºC and HR= 48%) for the seed to reach a balance point of 11% humidity andthus to preserve rice germplasm for periods of, at least, 5 years. Its capacity is forsome 14,000 entries stored in containers of 350 to 500 gr/entry.

Crosses: From the end of 1995, when FLAR was created, to the end of last year,and due to demands by partners, a total of 2,028 triple crossings (500 crossings/year) have been made. For the tropical zone 1,507 were done, for the following:Costa Rica (551), Colombia (327), Venezuela (583), Guatemala (41) and other(5). For the temperate zone a total of 521 crossings were done. Some of the latterwere processed by anther culture and R3 seed of some lines have been includedin VIOFLARs that have been distributed to the Southern Cone.

Early in the year, FLAR breeders, with collaboration of Eng. Edgar Torres ofFUNDARROZ, Venezuela, programmed a total of 263 triple crosses for thetropical zone. By request from Takazi Ishiy, the FLAR temperate zone breeder, atotal of 115 triple crossings for such region were made for a grand total of 378

152

triple crosses. Their F1 seed will be planted in Palmira during October to obtain F2populations to be sown in Santa Rosa in April, 2001.

Quality: FLAR continues providing laboratory analysis for grain quality as aservice for associate countries. During 2000, the laboratory received about21,000 samples for different quality determinations. Among these, there wereevaluated 1,500 from FEDEARROZ; 8,800 from CIAT rice project; 2,600 fromseveral countries and the rest from FLAR breeding lines.

The goal is to obtain material with intermediate amilose and gelatinizationtemperature (similar to North American rices). In Palmira, we determined thatseveral samples had appropriate reactions and, selected 43 samples that weresent to the USA for reconfirmation. Results indicate a very good correspondenceamong 2 laboratories and these 2 excellent characteristics are present in ourindica materials.

Experiments on the behavior to White Belly of materials harvested in severallocalities (Montería, Saldaña and Villavicencio) are being carried out. Montería isa good Hot Spot for White Belly and for lodging. All 21,000 lines were evaluatedfor WB and most of the material shows a consistent behavior accross locations.

A total of 310 F5 lines, were evaluated for their milling quality when they wereharvested in the correct time and also for their reaction to a harvest that wasdelayed for an additional 2 weeks. Of the 200 lines included in the VIOFLAR,77% maintain its milling quality when exposed to delays in harvest.

Rice Hoja Blanca Virus and Tagosodes: Rice Hoja Blanca Virus disease isendemic in the Latin America tropics and in the Caribbean. Integratedmanagement with tolerant varieties represents the best method for its control. Wecontinue, together with CIAT’s scientists, selecting, from the breeding program,tolerant lines to the virus and to its vector insect.

Table 3.5 is a summary of results obtained in F3 and F4 lines evaluated in 2consecutive semesters; the tolerant ones are incorporated in the VIOFLARs.

Using a methodology with Bluebonnet 50 as the susceptible check to evaluatemechanical damage to Tagosodes more than 2,000 F4 lines and more than 300 F5lines were evaluated. Over 50% of the lines showed resistance.

With CIAT’ scientists, a study to determine resistance expression of the Rice HojaBlanca (VHB), at different rice seedlings (table 3.6) was accomplished. Resultsdemonstrate that the new tropical genotypes are more resistant to HB than theprincipal resistant donor (variety Colombia 1). This variety is frequently used bothin our field and greenhouse evaluations. We now have natural resistance in all thecrop cycle and, therefore, are eliminating the need for the use of insecticides.

Low temperatures: Low temperatures limit rice production in the Southern Cone

153

and in some Caribbean zones. This year and according to Dr. Jennings’observations in his field trip to South, it was concluded that the methodology weare executing from Colombia (CIAT and FLAR) is not satisfactory. The mainreason is that we had never evaluated progenitors for cold tolerance that are usedin the crossing program. Cold being the main constraint and considering thatvarieties from Uruguay and Chile have excellent cold resistance, the following ofgermplasm flow was proposed in the last Temperate Zone Technical CommitteeMeeting, carried out in Cachoerinha on June 20-21,2000:

2. Obtain materials from the Southern Cone and evaluate it in CIAT undercontrolled conditions. These are more efficient than evaluations in the field,since there is much variability in the pressure under field conditions.

3. Recombine such resistance with the high yields observed in the tropicalmaterial of the VIOFLARs.

4. Make triple crosses and process them through anther culture. R2s areobtained in Colombia to harvest seed and dispatch it to Uruguay, Argentinaand IRGA.

5. Evaluate progenitors in Treinta y Tres, Uruguay.

In summary, this activity would consist in: evaluation of germplasm from Uruguay,Rio Grande do Sul and USA, for cold tolerance; 3 way crosses in Colombia;anther Culture and R2 in Colombia, evaluating blast and quality (part of the seed iskept and the rest is sent to the South); and R3 in Uruguay (an estimate of some10,000 lines). FLAR breeder selects the best lines for cold resistance andagronomic traits to form a VIOFLAR to be distributed to FLAR members. Duringwinter, evaluation of quality and blast resistance of R3 selected lines is done inColombia. Members cultivate VIOFLAR R4. Chile is a special case with climatesimilar to California and we could work with mutations and make crosses withLemont and then make back crosses to commercial varieties.

To begin cold characterization under controlled conditions, we have already atFLAR headquarters materials from Uruguay, IRGA, Chile and USA that werereported above (see "Exchanges with other sources").

For the next campaign in the Southern Cone materials were sent as it wasindicated previously (F2, F3 and material originating from anther culture), followingthe recommendations of the technical committee.

Anther culture: A total of 63 R2 lines of 15 different crosses are being evaluated todiseases in Santa Rosa during this year. The best ones will be evaluated to WB,TG, length of grain and amilose content. Then, selected R3 seed were sent toIRGA together with 100 lines R2 (45 crossings) selected and harvested in Palmiralast July.

Iron toxicity: We continue to rely on the methodology developed in Itajai, SantaCatarina, Brazil) to evaluate our progenitors and advanced lines.

154

Germplasm dispatch, year 2000:

Temperate-VIOFLAR 1999-2000 was formed with a total of 171 F5 lines derivedfrom 42 different crossings. Of the total, 21 lines are derived from anther cultureand the rest are lines from FLAR. Countries and number of distributed sets aredetailed in table 3.7.

Tropic-VIOFLAR 2000 was formed with a total of 200 F5 lines F5 (all FL) derivedfrom 44 different crossings. Countries and number of distributed sets areindicated in table 3.8. Preliminary results have indicated the following selections:

Calabozo, Venezuela = 61 lines (30,5%)Saavedra and San Juan of Yapacaní, Bolivia = 31 lines (15,5%)

Observations in Montería at 72 days, indicate that about a 15%% of lines behavewith a vigor similar to Fedearroz 50 and have a vegetative cycle betweenORYZICA 1 and Fedearroz 50, varieties adapted and planted commercially in thezone with 20 and 80% of the area, respectively.

Milling and White Belly tests were made with seed harvested in Saldaña(Colombia, a Fedearroz station). Data were compared with those obtained inPalmira and indicated a high congruency. This is of great importance since mostof lines included in this VIOFLAR show good percentages of whole rice, low WhiteBelly and tolerate a simulated 2 weeks delay in harvest without reduction of millingquality.

Seed multiplication: In Palmira we continue seed production of different blastsusceptible varieties, to conform the sprayers mixture used in the methodology toevaluate different genetic rice materials to such disease. Table 3.9 includesvarieties and quantity of available seed for the planting in Santa Rosa on 2001.

VIOFLAR 1999 Results: tables 3.10 and 3.11 indicate number of distributed sets,data received, as well as number and percentage of selection in the germplasmincluded in Temperate and Tropic- VIOFLAR, respectively. In Itajaí, SantaCatarina, Brazil, 12 lines were selected for their good behavior for yield trials inthe cycle 1999-2000. In 2 localitions in Bolivia a total of 26 lines were selected.In Uruguay 36 lines were selected in the cycle1999-2000, reporting yieldsbetween 12 and 13 t/ha in Artigas; the cold is not a stress at that site.

Several lines from the cross FL00144 (CT8008-16-31-3P-M// CT9682-2-M-14-1-M-1/ CT11008-12-3-1M-4P-4) and line FL0007-17P-12-2P-M (IRGA 416/CT10865-AC-12-M// CT8008-16-31-3P-M) had notable perfromance in all 3countries. In Uruguay, an anther culture line FL00220-AC-5-M (CT8008-16-31-3P-M/ CT10865-AC-12-M// IRGA284-18-2-2-2) showed good performance.

For Tropic- VIOFLAR (table 3.11) results supplied by partners indicated that inCosta Rica, 2 of 45 selected lines (FL00159-6P-2-1P-M and FL00159-6P-2-9P-M)are in Regional Yield Trials year 2000, and the others went to Observation Plots

155

during this same year.

Guatemala reported a total of 30 lines out of which 20 are in yield trials this year, 8are used for crossings and the other 2 for both purposes.

Nicaragua, based on 2 localities, reported 20 promissory lines, which are indifferent tests this year. In Venezuela, 7 lines were submitted to Preliminary Trialsof the National Plan, where FL00147-8P-15-5P-M (CT8008-16-31-3P-M//CT9682-2-M-14-1-M-1-3P-M-1/ CT10310-15-3-2P-4-3) has stood out for its highyields. That line was also selected in Costa Rica, Nicaragua and Guatemala.

New Varieties: We keep as much as possible an updated database concerningvarietals designations (table 3.12). Twelve releases took place by FLAR membersthis year.

Two varieties were released in Venezuela; they were introduced through INGER-LAC nurseries. Line CT10310-15-3-2P-4-3 released as Fundarroz PN-1,represents a genotype with genetic diversity since it originates from a triplecrossing made at 1988 in CIAT in 1988 where indica material x Japónica (P 3084-F4-56-2-2/ ITA306// CT8154-1-9-2) was combined. The other variety is lineCT8240-1-3-9P-M that originates of triple crossing, done also in CIAT in 1986 (P5446-6-3-2/ CT5690-3-19-2// P 3059-F4-79-1-1B). This one was released asFONAIAP 2000.

In Panama, the Agricultural Research Institute of Panama (IDIAP) officiallyreleased a new rice variety called IDIAP L-7 for irrigation ecosystems and favoreduplands. It corresponds to a selection of line CT8008-3-5-9P-M (released byUniversity of Panama, in that same country, as Panama 3189), that wasintroduced through a Observational Rice Nursery for Latin America (VIOAL 1989).Its current pedigree is CT8008-3-5-9P-M-RH7. Thus, Panama has 2 varietiesoriginating from the same crossing and this is the tenth variety registered in ourdatabase from cross CT8008, made in 1986.

In Colombia, FEDEARROZ has been approved to release 4 new varieties:Fedearroz 2000, Colombia XXI, Fedearroz La Victoria 1 and Fedearroz LaVictoria 2.

Fedearroz 2000 corresponds to line CT10323-29-4-1-1T-2P originating from triplecrossing P 3084-F4-56-2-2/ P 3844-F3-19-1-1B-1X// CT8154-1-9-2 made in 1988.This variety differs from Fundarroz PN-1 only in the male progenitor of the simplecrossing. Variety Colombia XXI is line FB0100-10-1-M originating from simplecrossing P 5413-8-5--11/ CT9145-4-15-1-1 accomplished in Bosconia byFedearroz. Fedearroz La Victoria 1 is line CT10240-10-1-2-1T-2-1 result of thetriple crossing CT6129-17-7-9-1/ P 4278-F2-84-1-1X// C 48CU76-3-2-1-4-5Maccomplished in CIAT in 1988. And Fedearroz La Victoria 2 corresponds to lineCT10192-5-1-2-2T-2-1 originating from the triple crossing CT7415-6-5-3-2X/Ceysvoni// CT8163-9-4-4, accomplished also in the CIAT in 1988. Fedearroz

156

2000 has been released for the center of the country; Colombia XXI for the dryCaribbean and La Victoria 1 and 2 for upland conditions in Humid Caribbean.

Other named varieties include 4 in Brazil: Rio Grande introduced as P 4725-F2-9-1, and varieties denominated IRGA 418, 419 and 420 of local crossings. Ecuadorreports a new variety, INIAP 14, originating from introduced material throughINGER nurseries as PSB RC-12.

As a means of preliminary information, based on data of leaf blast, year 2000, wehave: out of 2,693 F2 tropic populations, represented in 268 crossings, 46%(1,239 populations) have been eliminated for its susceptibility to leaf blast. In thesame way, out of the 86 F2-Temperate originated of 99 crossings, 58,4% (342populations) has been eliminated. For the 2,186 F4-Tropic lines originated with108 crossings, we find a 35,1% of susceptibility (768 lines).

SOME OBSERVATIONS ON BLAST. Year 2000.

Analyzing field observations and leaf blast records, we have found someanomalies:

--Of 268 crossings originating the F2 population, 189 were made for Venezuelausing the Lineage Exclusion concept. These 189 crossings include 2,017families, out of which 737 are susceptible to leaf blast, and 1,280 were resistant,or segregants. Neck reaction is pending as we want to further analyze differentprogenitors involved in susceptible crossings, and to try to establish what ishappening.--Of 2,186 F4 lines, 768 are susceptible to leaf blast. These susceptible lines wereoriginated mostly in F2 populations that had evaluations of 1.2 or 3 last year.

As an example we can analyze the following crossing:

FL01869 include 128 lines: 79 susceptible and 49 resistant.Parent 1: IR841-63-5-1B, highly resistant to 6 lineages.Parent 2: CT11424-14-F4-12P-1P, highly resistant to 6 lineages.Parent 3: CT9509-17-3-1-1-M-1-3P-M-1, highly resistant to 5 lineages andsusceptible to one of them. At field level, Parent 1 resulted susceptible to leafblast and the other 2 parents were resistant.

--Of the 256 progenitors most frequently used to develop new rice genotypes wehave that:Ninety nine are highly resistant (HR) or Resistant (R) to 6 lineages and showedresistance in field to leaf blast last year. Of these, 49 have been characterized asHR to all lineages and also field resistant to Blast last year. Now, 27 (55,1%) ofthem are susceptible in field and 22 maintain their resistance. Of the 50progenitors that are HR or R and with resistance to BL last year, 39 (78%)resulted susceptible and 11 resistant in field.

157

--Similarly, a 32.5% of leaf blast susceptibility in VIOFLAR 2000’s lines has beenobserved. This is worrying, since these lines originate of selections from in F2 andF4 generations that was resistant to blast in leaf and neck. Progenitors used insome crossings that have contrasting reactions have been analyzed and it isobserved that in all of them lineages exclusion was applied.

3.A.2.2. Elaborate breeding plan for the Temperate region

Milestone: the plan is ready (see 4.1.1 above) and will be submitted for approvalby the Administrative Committee of FLAR in November 2000.

3.A.2.3. Evaluate VIOFLARs in all member countries and share results withall members (see 4.2.1 above).

3.A.2.4. Organize and implement Breeders workshops:

Milestones:- Cachoeirinha, Brazil: April 2000- Santa Rosa, Colombia: August 2000.

Breeders workshop, temperate zone, Cachoeirinha, April 2000. A total of 1,423lines were selected and seed was dispatched as follows: Argentina Entre Ríos(356); Argentina Corrientes (120); Uruguay (299); EPAGRI (674); IRGA (102).Argentina was the only member that had completed evaluation of the materialsfrom the VIOFLAR 1999. The results show relatively low rates of selection: 20 to25 % at El Encuentro, 14 % at Ita Caabó and 25 to 30 % at El Rocío. In SantaCatarina, the control variety was superior to every line in the set. All lines areearlier than the variety. The best lines were quite comparable to the check: FL306, 227 and 482. Grain discoloration was a major problem both in Argentina andBrazil.

Breeders workshop, tropical region, Santa Rosa, August, 2000. All members fromthe tropical region were present: Colombia, Cuba, Costa Rica, Guatemala,Nicaragua, Panama, and Venezuela, plus Brazil and Bolivia, as well as CIAT.

Due to blast susceptibility, 46% of F2 families and 45% of F4 families werediscarded. These numbers are similar to those of previous years. It is somewhatalarming that, in VIOFLAR 2000, 45% were susceptible compared with last yearwhen all lines were resistant. Perhaps even worse, out of 99 progenitors that wereincluded as resistant in previous years, 65 of them appeared as being susceptiblethis year.

The results of selection by researchers at the workshop show that:

--Population F2: Out of 2,693 F2 families (obtained from 268 crosses), 246 wereselected (9.1%), and 1,323 individual plants were harvested (F3 seed). These F3lines, represented by 85 crosses (31.7% of total crosses), will be evaluated to

158

White Belly (WB), gelatinization temperature, length, Hoja Blanca, Tagosodes,and will be planted in Monteria (Colombia) for evaluation to lodging and WB aswell as in Pamira to obtain F4 seed.

--F4 lines: out of 2,186 lines F4 (108 crosses), 238 lines (10.9%) were preselectedin the field, corresponding to 66 crosses (66%). These preselcted lines will beevaluted for WB, TG , L, Amylose (seed form Santa Rosa) and to HB, Tagosodesand milling quality from seed from Palmira.

159

Table 3.1. Number of sets distributed and data received for VIOAL 1999.INGER-LAC.

Country No. Sets VIOAL No. Sets VIOAL-Acid S..

Distributed Received Distributed ReceivedArgentina -- -- 1 1Bolivia -- -- 1 0Brazil 1 1 -- --Colombia 1 1 1 0Cuba 1 0 -- --Ecuador 1 0 -- --Guatemala 3 0 1 0Nicaragua 4 2 -- --Panama 3 2 1 0Peru 2 0 -- --Venezuela 2 1 -- --TOTAL 18 7 (38.9%) 5 1

VIOAL = 130 Lines VIOAL-Acid Soil = 28 Lines

DATA RECEIVED VIOAL:

Country/Site Selected Lines %

Nicaragua: Malacatoya Sebaco

2635

20.026.9

Panama: Tocumen Río Hato

5022

38.516.9

Brasil – Itajaí 14 10.8

Venezuela-Barinas -- --

Total Others 77 59.2

160

Table 3.2. Number of Sets, VIOAL 2000, for Latin America distributed, firstsemester, 2000

COUNTRY NO. SETS DISTRIBUTED

Costa Rica 3El SalvadorPanamaColombia: CORPOICA

13

2 Fedearroz AventisVenezuelaArgentinaEcuador

11221

TOTAL 16

VIOAL= 64 Lines.

Table 3.3. Rice Germplasm planted in Villavicencio and Palmira, 2000 A.

Class of material No. Lines No. Crossings-Germplasm Bank (BCF) 256 178-VIOFLAR: Tropic, 2000 207 44- F4 Lines : Tropic 2186 108-F2 Populations: Tropic Temperate

2693 586

268 99

-F1 triple crosses -- 72-R2 (Anther Culture) 63 15TOTAL 5991 784

161

Table 3.4. Rice Introductions planted in Villavicencio and Palmira. 2000 A

Class of material No. Lines No. Crosses

VIOAL, 2000 67 57

- Countries: Peru Tailandia, 1998 Venezuela

11 151 88

4 10 35

- CIAT: F4 Lines (Altitude Project) Lines from Inter Specific crossings Anther Culture F2BC2 from Bg90-2/O.glaberrima-INGER-LAC (VIOAL, 97, 99)

2041,554 268 61 26

26124 73 1 21

Total 2,430 351

Table 3.5. Results of evaluations to Rice Hoja Blanca Virus in F3 and F4Lines for the tropics. FLAR, 1998B al 2000A

Reaction (1-9)Class of material Total lines Tolerant

(1-3)Intermediate

(5)Susceptible

(7-9)

F3 (1998B) 9,172 6,572 (71.7%) 1,478 (16.1%) 1,122 (12.2%)

F4 (1999A) 3,738* 2,570 (68.8%) 710 (19.0%) 458 (12.2%)

F3 (1999B) 3,000 1,975 (65.8%) 468 (15.4%) 563 (18.8%)

F4 (2000A) 2,186** 1,144 (52.3%) 294 (13.5%) 748 (34.2%)

TOTAL F3

TOTAL F4

12,172

5,924

8,547 (70.2%)

3,714 (62.7%)

1,946 (16.0%)

1,004 (16.9%)

1,685 (13.8%)

1,206 (20.4%)

* Out of the F4 Lines, 200 F5 Lines were included in VIOFLAR, 2000.** F5 Lines from this group will form VIOFLAR 2001 for tropical countries.

162

Table 3.6. Expression of resistance to Rice Hoja Blanca Virus in riceseedlings. CIAT 1999-2000

Percentage of resistant plants

Material 5 dae* 10 dae* 15 dae* 20 dae*

I F I F I F I F

Fedearroz 2000

Fedearroz Victoria 1

Fundarroz PN1

Fedearroz 50

Colombia 1

Bluebonnet 50

45 56

25 54

35 29

5 17

5 33

5 15

60 88

35 88

55 83

50 72

35 67

0 15

40 94

50 95

30 73

35 55

30 73

10 17

85 88

50 91

65 71

50 48

40 72

0 15

* dae= days after emergence; I = greenhouse infestation; F = field infestation.

Table 3.7. Number VIOFLAR Sets 1999-2000 for the Temperate Zonedistributed to FLAR Associates. October, 1999

Country No. Distributed Sets

Cuba 1Brazil• IRGA 1• ITAJAÍ 1Chile 1Uruguay 2Argentina 1

Total 7

VIOFLAR-Temperate = 171 Lines.

163

Table 3.8. Number of Temperate Zone VIOFLAR 2000 Sets distributed toFLAR associates. January –February, 2000

Country No. Distributed Sets

Bolivia 2Colombia 4Costa Rica 2Cuba 1Guatemala 2Nicaragua 1Panamá 2Venezuela 2

Total 17

Table 3.9. Seed production of commercial varieties to be used asspreaders in Santa Rosa, Villavicencio, Colombia. Palmira. 2000 B

Bags TotalVariety 1998 1999 2000 Bags Kg.

1. Orizica 1 -- 2.5 19.2 21.7 10852. Cica 9 4.5 -- 17.0 21.5 10753. Linea 2 -- -- 19.0 19.0 9504. Oryzica Caribe 8 -- -- 17.0 17.0 8505. Selecta 3-20 -- -- 18.5 18.5 9256. Cica 8 -- 3.0 2.0 5.0 2507. Metica 1 10.0 -- -- 10.0 5008. Fanny (Irri) -- -- 24.0 24.0 1200

Total 176.7 8835

164

Table 3.10. Number of Temperate Zone VIOFLAR 1998-1999 Sets distributedin October, 1998. FLAR

Country No. SetsDistributed

DataReceived

Selected LinesNo. %

Brazil: IRGA ITAJAI

42

01

--12

--9.1

Bolivia 2 2 26 19.7Cuba 1 0 -- --Paraguay 1 0 -- --Uruguay 2 2 36 27.3Total 12 5 (41.7%) 58 * 43.9

VIOFLAR-Temperate = 132 Lines.• =Total Different Lines.

Table 3.11. Number Of Tropical Zone VIOFLAR 1999 Sets Distributed DuringJanuary-March, 1999. FLAR.

Country No. SetsDistributed

DataReceived

Selected LinesNo. %

Colombia 1 1 -- --Costa Rica 2 1 45 66.2Guatemala 2 1 30 44.1Nicaragua 2 2 20 29.4Venezuela 2 2 7 10.3Panama 1 0 -- --Total 10 7 (70%) 55* 80.9

VIOFLAR-Tropic= 68 Lines.*= Total of different Lines.

165

Table 3.12. Rice Varieties released in 1999 - 2000

No. Name Pedigree Crosses Country1 RIO GRANDE P 4725-F2-9-1 P 2026-F4-49-5-5//IR5533-13-1-1/ORYZICA 1 Brazil2 IRGA 418 IRGA 284-1-18-2-2-2 BR-IRGA 412/CICA 9//BR-IRGA 409 Brazil3 IRGA 419 IRGA 369-31-2-3F-A1-1 ORYZICA 1/BR-IRGA 409 Brazil4 IRGA 420 IRGA 370-42-1-1F-C1 ORYZICA 1/BR-IRGA 412 Brazil5 FEDEARROZ 2000 CT10323-29-4-1-1T-2P P 3084-F4-56-2-2/P 3844-F3-19-1-1B-1X//CT8154-1-

9-2Colombia

6 COLOMBIA XXI FB0100-10-1-M-1-M P 5413-8-3-5-11/CT9145-4-15-1-1 Colombia7 FEDEARROZ LA

VICTORIA 1CT10240-10-1-2-1T-2-1 CT6129-17-7-9-1/P 4278-F2-84-1-1X//C48CU76-3-2-

1-4-5MColombia

8 FEDEARROZ LAVICTORIA 2

CT10192-5-1-2-2T-2-1 CT7415-6-5-3-2X/CESYSVONI//CT8163-9-4-4 Colombia

9 INIAP 14 PSB RC-12 Ecuador10 IDIAP L-7 CT8008-3-5-9P-M-RH7 P 3050-F4-52/ORYZICA 1//IR21015-72-3-3-3-1 Panama11 FONAIAP 2000 CT8240-1-3-9P-M P 5446-6-3-2/CT5690-3-19-2//P 3059-F4-79-1-1B Venezuela12 FUNDARROZ PN-1 CT10310-15-3-2P-4-3 P 3084-F4-56-2-2/ITA 306//CT8154-1-9-2 Venezuela

166

3.A.3. Crop ManagementEdward Pulver, Crop Management Specialist, FLAR (consultant)

Luis R. Sanint, Rice Economist and Executive Director, FLAR

Collaborators: twelve FLAR member countries, CIAT, IRRI and CIRAD.

3.A.3.1. Collaborate in preparation/implementation of Agronomy plans

3.A.3.2. Collaborate in surveys of crop production constraints

Milestones: quick overview of crop management practices. Prepared a projectprofile that was presented at FAO International Rice Commission, September2000, Rome.

3.A.3.2.1. Introduction

Rice production in the Latin America and Caribbean (LAC) region in 1999surpassed 24 million MT (paddy). This record harvest is the cumulative result ofyears of improvement in yield, which accounts for all the production increasesrecorded in the LAC for the last two decades. During the 1980s, rice yieldsincreased at an annual rate of 3.0%, primarily due to the expansion of highyielding varieties in the irrigated production system. During the 1990s, advancesin yield grew by 3.5%/annum (table 3.13), which is a reflection of furtherimprovements in irrigated rice and a decline in the low-yielding upland rice sector.

The rapid and consistent advances in yield of irrigated rice and the decreasedemphasis on unstable upland rice have resulted in an increased interest inproduction under irrigation. Historically, upland rice production has accounted for asignificant portion of total rice production, especially in Brazil and Central America.However, the trend toward free-markets and the continued improvements in irrigatedrice productivity have exposed the upland sector to free market forces andretractions in government support. The end result has been a significant decline inupland production and the emergence of irrigated rice as the dominating riceproduction system.

The rapid advances witnessed in irrigated rice production during the last 20 yearshave been primarily a result of variety improvement. Currently, high yielding varietiesoccupy 90% of the irrigated rice production area. However, yields remain far belowthe genetic potential of available varieties. National yields of irrigated rice seldomsurpass 5 MT/ha, even though the yield potential of current varieties surpasses 8MT/ha. Additionally, yields recorded from experimental plots and in fields ofprogressive farmers are often 50% greater than regional or national yields. The largevariation in yield even within a homogenous production zone is due to differences incrop management practices.

The difference between potential yield and actual farmers' yield is referred here asthe "yield-gap". Bridging the yield-gap represents the most immediate opportunity for

167

increasing yields. Factors contributing to the yield-gap vary considerably betweenproduction systems, ecological zones and economic situations. Although, there aresome universal constraints to improved yields, such as inefficient weed control andcrop fertilization, most constraints are localized. Consequently, efforts to bridge theyield-gap must focus at a national level and, if feasible within a production ecology.

3.A.3.4. Assessing the Yield-Gap

This report assesses the yield-gap in the12 LAC countries that are members ofFLAR. Collectively these 12 countries account for approximately 70% of all irrigatedrice production in the LAC. All the selected countries have established a nationalgrower association. This organizational structure facilitates a means of addressingthe yield-gap problem and also provides a mechanism for conducting on-farmresearch and technology transfer, an essential ingredient for improving cropmanagement.

The 12 targeted countries consist of five in the Southern Cone of South America,including Argentina, two states in southern Brazil, Bolivia, Chile and Uruguay. Due tothe geographic location and ecological conditions, the five countries are grouped intoa Temperate Region.

There are seven countries located in tropical South and Central America and theCaribbean country of Cuba. The tropical countries have similar productionconstraints and are organized into a Tropical Region.

3.A.3.4.1. Temperate Region

• Brazil

Brazil is the most important rice producing country in the Americas with an annualproduction of approximately 11 million MT. Traditionally, production has been evenlydivided between the highly productive irrigated areas of the south and the vastextensions of upland rice in the Cerrados. However, during the last 10 years,production from the unstable upland sector has declined considerably. In 1985/86Brazil had approximately 4.5 million hectares in upland production that accounted formore than 50% of national production. In 1996/97, the area planted to upland rice inBrazil declined to 2.3 million hectares and contributed only about one-third ofnational production. Data presented in table 3.14 provides a summary of the trendsin production during the last 10 years. National yields have increased at a growthrate of 4.25% while area cultivated to rice decreased at an annual rate of 2.86%,resulting in an annual growth rate in production of 1.26%. This information reflectsthe decline in upland rice and the continuous growth in irrigated rice.

The two most important irrigated rice-producing areas in Brazil are Rio Grande doSul (RS) and Santa Catarina (SC). Although these two states are neighbors, theyhave distinct production systems. Relatively large farms with most of the plantingsconducted in the traditional dry-land preparation and seeding characterize

168

production in the RS. In contrast, the farms in SC are small with nearly all theproduction occurring in puddled soils and water-planted using pre-germinated seeds.The agronomic constraints contributing to the yield-gap are distinct in these twogeographically similar areas.

Brazil - Rio Grande do Sul

The Rio Grande do Sul (RS) is the pioneer area for irrigated rice in Brazil. Cultivationof irrigated rice commenced in the early 1900s but production grew most rapidlyduring the 1970s and 1980s following the introduction of high yielding, semi-dwarfplant types. In 1970, approximately 400,000 ha of irrigated rice were grown with anaverage yield of 4 MT/ha. In 1980, the area increased to 500,000 ha but the yieldremained constant at 4 MT/ha. In 1990 the area continued to expand and reached800,000 ha with an average yield of 5 MT/ha. However, during the last 10 years thearea has remained relatively constant and yields have stagnated at around 5 MT/ha.In general terms, the yield increases observed during the 1970s and 80s providedthe economic stimulus for further expansion of irrigated area. In contrast, yieldstagnation during the 1990s has been the primary limitation for further expandingirrigated area. Numerous studies have reported that RS has suitable land and waterresources to support a minimum of 2 million ha of irrigated rice. Consequently,increasing the yield of current production area can provide the required economicincentive for further expansion of irrigated rice production.

Research and extension in RS is provided mainly by the Instituto Rio Grandese doArroz (IRGA). IRGA is supported exclusively by production check-offs. Research isconducted at a central research and several small satellite stations. Most of theextension work involves providing services to growers, especially in the area ofirrigation management. Extension is in a transition period in which increasedemphasis will be given to technology transfer activities in area of crop management.

The state is divided into six production zones that are distinct in terms of topography,soil fertility, microclimates and crop management. In summary, information providedin table 3.15 illustrates that the yield-gap across all six-production zones in RS is 1.3MT/ha. Technology for bridging this yield-gap is readily available but has not beenextended to many growers. Simply bridging the yield-gap would increase theaverage yield in RS from its current level of 5.2 MT/ha to 6.5. This would result in anadditional production of nearly 1 million MT over the entire state.

Brazil - Santa Catarina (SC)

The state of Santa Catarina has approximately 120,000 ha of irrigated rice, mostly insmall, family farms of less than 15 ha. Almost all preparation and planting is in waterusing pre-germinated seeds. The state average yield is 5.7 MT/ha but yields varygreatly between production zones and between farmers within a given productionarea. Historically, the state suffered from high infestations of red rice but thecontinuous use of puddling and pre-germinated seeds has significantly reduced theincidence of red rice. The state has a very organized seed industry that produces the

169

highest quality rice seeds in Brazil. However, use of certified rice seed in SC islimited to only about 30% of the growers and this is a major constraint to improvedproduction. Due to the low demand locally for certified rice seed, the seed industry iscurrently producing rice seeds on contract for other areas in Brazil. A very effectivestate-support research program (EPAGRI) supports the rice industry. Prior to 1990,rice research and extension were combined into one program that was instrumentalin increasing productivity to over 5.5 MT/ha - the highest state average in Brazil.However, reorganization of the state research and extension resulted in separatingthe two agencies and the end result has been a sharp decline in rice extensionactivities and slow progress in yield improvement.

There are six major rice-producing regions in SC. Information for current area, yield,production and potential production for each production zone is provided in table3.16. The productive zones, for example the Alto Vale do Itajai, are relatively newand still have a relatively high level of indigenous soil fertilizer and are least infestedwith red rice. In contrast, the least productive region is the Litoral Centro, which isthe oldest production zone. State average yield is 5.7 MT/ha but the yield variesfrom 7 MT/ha in Alto Vale do Itajai to only 4.7 MT/ha in the Litoral Centrol.

The estimated yield-gap for the entire state is 1.2 MT/ha. This is the yield-gap thatcan be bridged in the short-term, since much of the technology for increasingproductivity is readily available and only needs to be extended to the growers. Thistechnology includes use of red-rice free seeds, appropriate timing with adequatedoses of fertilizer and good water management. A state average yield of 6.9 MT/hais readily obtained by bridging the yield-gap, resulting in an increase in production ofnearly 140,000 MT (table 3.16).

• Argentina

Argentina is not a traditional rice producing country, even though the country hasvast areas that are highly suited for irrigated rice production. In the early 1980s,Argentina cultivated only 80,000 ha, with an average yield of 3.2 MT/ha, and a totalproduction of approximately 250,000 MT. Rice was viewed as a secondary crop withlittle economic importance due to low internal demand and limited opportunities forexport. Rice production was viewed as a "low-input, low-output crop". Most of theavailable varieties were long-grained US-bred varieties, that were selected mainlyfor grain quality that enabled domestic production to compete in terms of quality withimported rice from the US. Few farmers used inputs, only minimum fertilizer wasused and few used chemical weed control products. During the 1990s theMERCOSUR Trade Agreement was implemented, permitting the export of rice tomember-countries without restrictive tariffs. As a result of available markets, riceproduction in Argentina expanded tremendously since then, with area growing at anannual rate of nearly 12%, yields at an annual growth rate of 2.8% and, therefore,total production increased at an annual rate of nearly 15% (table 3.17). During thelate 1990s, yields averaged in excess of 5.5 MT/ha and total production surpassed1.6 million MT. However, devaluation of the Brazilian real and growing subsidies inthe world market were responsible for a rapid contraction in paddy rice production; it

170

fell from 1.6 million tons in1999 and 280,000 has, to 800,000 tons and 150,000 hasin 2000 and production in 2001 will be around 400,000 tons (in about 75,000 has), aset back to the level of production of the mid 1980's.

Rice production in Argentina has great potential. There are vast land and waterresources available and production is competitive within the region. More immediateopportunities for increasing production exist by improving yields. The yield potentialin most of the production areas is in excess of 7 MT/ha due to highly favoredecological and environmental conditions, such as good soils, adequate water, coolnight temperatures, high light intensity and long days during the growing season.Furthermore, Argentina is developing varieties, via its association with FLAR, thatare more adapted to local growing conditions. The key to further advances in yield isincreased use and improved timing of essential inputs. The rice industry, in its urgeto remain competitive, has altered farmers' perception of rice from a low-input cropto an income-generating activity.

The current yield-gap is nearly 2 MT/ha. In the long term, bridging the yield-gap willpermit more competitive production permitting the country to increase exports andprovide the economic stimulus for expanding the area under production.

• Uruguay

Irrigated rice production in Uruguay has nearly tripled during the last decade. Duringthe decade of the 1990s, production grew at an annual rate of over 11%, due togrowth in area (8.5% annual growth rate) and improved yield (2.5% annual growthrate). In 1999, rice production approached 1 million MT (table 3.18).

Uruguay has the highest yields in the LAC, with a national average of 6.0 MT/ha.However, the yield potential is in excess of 7 MT/ha. The environmental andecological conditions in Uruguay are similar to Argentina and favor high productivity.Nationwide the yield-gap is estimated at 1.0 MT/ha. Uruguay has a well-establishedrice industry and increased focus on crop management can produce results in arelatively short period. Due to existing export market in Brazil, growers have aneconomic incentive for more competitive production.

• Bolivia

Rice production is relatively new in Bolivia. All production is located in the sierraarea, around Santa Cruz. Agriculture development in the region accelerated duringthe last two decades as a product of World Bank assistance. Much of the area isprime agriculture land and is used for upland rice, soybean and pasture production.Most of the rice production is rainfed and depending upon soil types, production isclassified as either favored or unfavored upland.

There are approximately 120,000 ha in production. Approximately, 70,000 ha arecultivated by large growers (greater than 500 ha), in which upland rice is rotated withsoybean and/or pastures. Another 20,000 ha of upland rice are farmed by a

171

cooperative of Japanese-Bolivianos near the Colonial San Juan. The remaining30,000 ha belong to small growers (about 17,000 rice growers out of 20,00 in thecountry), many of which practice slash/burn, shifting cultivation. More than 50% ofthe small farmers are located in marginal areas, where soil problems limitproduction. In addition, there are approximately 200 ha of favored upland production,where supplemental irrigation is applied.

Due to the dependence of production on rainfall, yields are unstable and low.Productivity is estimated at 3 MT/ha in the favored, lowland rainfed areas and 1MT/ha in the unfavored upland, where most of the small growers are located. Therehas been very little progress in yield improvement or expansion of production areaduring the last decade (table 3.19).

The last two seasons have been very difficult for rice growers due to drought. As aconsequence, the organization comprising the large growers, the cooperative fromSan Juan and the small-growers association united to form a national rice growersassociation. The purpose of the unification was to join FLAR with the intent ofgaining access to irrigated rice germplasm and technical assistance for converting toirrigated rice production. The national association is initiating pilot irrigation projectsas a means of demonstrating to growers irrigated rice production technology. Thesierra area has abundant land with underground water and the area is highly suitedfor irrigated rice production.

In the case of Bolivia, the yield-gap was determined by estimating the differencebetween a feasible yield under irrigation as compared to current yield under uplandproduction. A reasonable yield under irrigation is estimated at 5 MT/ha, although theyield potential under irrigation is much greater. Current yields under favored uplandare approximately 3 MT/ha, resulting in estimated yield-gap of 2 MT/ha. This yield-gap is estimated for 10,000 ha, which is a reasonable area to convert to irrigation inthe short-term. Consequently, yield improvement on 10,000 ha will result in aproduction increase of approximately 20,000 MT.

• Chile

Chile is not a major rice producer, even though environmental conditions arefavorable for very high productivity, similar to California. Chile normally cultivatesless than 30,000 of irrigated rice, with an annual production of 100,000 MT. Averageyield is less than 4 MT/ha, due to poor water and general crop management.Information presented in table 3.20 illustrate there has been very little progress inproduction in Chile over the last decade. The area planted to rice is on the declineand national yields have remained stagnant.

The yield potential of irrigated rice in Chile is in excess of 8 MT/ha; however, thisyield level is feasible only on areas that have adequate drainage. In general, theyield potential is estimated at 6 MT/ha, resulting in a yield-gap of 2 MT/ha.

172

3.A.3.4.1.1. Summary of yield-gap in Temperate Region of LAC

The Temperate Zone of South America is a high-yield potential area due tofavorable climatic factors, and abundant water and land resources suitable forirrigated rice production. However, the average yield across the five countriesanalyzed is only 5 MT/ha. The yield potential for zone is estimated at 6.5 MT/ha,resulting in a regional yield-gap of 1.5 MT/ha (table 3.21).

Currently there are approximately 1.42 million ha of irrigated rice in the five countriesassessed with an annual production of 7.1 million MT. In the short-term simplybridging the readily apparent yield-gap would increase production to 9.3 million MT.This represents a 31% increase in production with a farmgate value of US $330million.

In the long term, the economic returns from bridging the current yield-gap are muchgreater. Bridging the readily apparent yield-gap would provide the economic stimulusfor expanding irrigated rice production since much of the increases in productivitycan be obtained by improving the crop management, which does not require largecapital investments. Most of the required agronomic practices involve use ofavailable inputs in a more timely fashion. Due to the difficult problem with red ricethat already exists in southern Brazil and the potential for the problem to become aserious constraint in other areas, special attention must be directed to preventiveprograms. The rapid expansion of new irrigated areas in Argentina andtransformation to irrigation in Bolivia are prime areas where strict measures shouldbe taken to prevent the introduction of red rice. Use of high quality seeds devoid ofweedy or red rices is the easiest and most economic means of reducing red riceinfestations.

3.A.3.4.2. Analysis of yield-gap in tropical South and Central America

• Colombia (for additional on Colombian rice yield gaps details, see 3.4.5below)

Rice is grown in Colombia under a series of systems, including favored upland,water preparation and seeding and convention dry land and dry seeding. All systemshave distinct management practices and constraints to improved production arehighly variable. Colombian growers have a history of indiscriminate use of pesticidesand excessive production costs. Concerted efforts by the national growerassociation (FEDEARROZ) over the last 15 years have resulted in reduced pesticideusage and declining production costs. However, despite an active farmer-supportedresearch and extension service, overall rice production in Colombia had beenstagnant for the last two decades. During the 1990s, rice production increased at anannual rate of only 0.1%, despite a 1.5% annual growth in yield. The main problemin Colombia has declining area, which is more of social problem than technicalissue. During 1999 and 2000, though, production bounced back to about 2.3 milliontons of paddy rice in 450,000 has, for an average rice production of 5 tons/ha. Thishas been the result of better varieties and government support to rice producers.

173

Rice yields in Colombia vary considerably between production regions. Informationprovided in table 3.22 shows that the average yield between production zones variesby 50%. However, it is apparent the yields are highly variable even within a relativelyhomogenous production zone. A recent report by INDUARROZ (Colombian RiceMillers Association) stated that improper timing and inadequate doses of herbicidesand fertilizer applications contribute significantly to the large fluctuations in yields.Also, dates of planting have major influence on final yield. All of these variables aremanagement practices and subject to improvement provided farmers are aware ofthe impact they exert on final yield.

Production information for the rice growing regions of Colombia accompanied bydata on the yield potential of each zone is presented in table 10. Nationwide theyield-gap is estimated at slightly less than 1 MT/ha but ranges from 1.3 MT/ha forthe major Centro production zone to only 0.4 MT/ha for the Llanos region, wheremost of the rice is rainfed.

Yields at the national level exhibit big gaps among groups of farmers. But thegaps are mostly explained by edapho-climatic conditions of each region and byproduction systems. Bridging the yield-gap between actual and potential yields ismainly a function of focusing the research activities on key problems anddemonstrating the improved management practices to the growers.

• Venezuela

Producers in Venezuela prepare land in water and seed pre-germinated seeds,which is an uncommon system in the Americas. This system has distinctmanagement requirements and presents unique problems. The major constraint toyield improvement is the release of toxins from anaerobic decomposition of organicdue to continuous flooding and puddling. This disorder is speculated to be similar tothe yield decline problem observed in Asia in the intensively cropped, continuouslyflooded production system.

During the 1990s, rice production in Venezuela experienced an annual growth rateof 2.9%; however, expansion of area under cultivation accounted for more than 80%of the increased production. National yields remain relatively low at 4.3 MT/ha andhave been essentially stagnant during the last decade (0.5% annual growth rate inyield). Yield stagnation is primarily a product of the increased incidence of the yield-limiting disorder associated with the release of toxins from long-term anaerobicconditions, referred to locally as the "black root syndrome".

There are two major rice-growing regions in Venezuela, each with distinct problems(table 3.233.23 In the Llanos Centrales, most production occurs within an irrigationdistinct feed by a large reservoir. Water supply limits production to one crop/year.The low average yield in this area of 3.8 MT/ha is a reflection of farmers growing thecrop out of season when water supply is limited. During the normal growing season,yields are higher and approach 6 MT/ha. Excessive pesticide use and inappropriate

174

timing of fertilizer applications, especially N, are the primary management problems.Managing N fertilizer in the puddling production system is difficult due to the largelosses experienced when applications are made on wet soil. It is feasible tosignificantly increase yields by applying all N on dry soil prior to land preparationand/or draining and applying N on dry soil during late stages of growth.

In the Llanos Occidentales, intensive rice production is the norm resulting in nearlycontinuously flooded soil. This area accounts for more than two-thirds of the nationalproduction and is heavy affected by the black root syndrome. Altering landpreparation systems between crops, which will permit the soil to dry between crops,can significantly increase yields. Altering land preparation or complete conversion todry land preparation appears to significantly reduce the incidence of the black rootdisorder and stimulates yield. The yield-gap in the Llanos Occidentales is estimatedat 1.6 MT/ha, which can be bridged by adopting improved land preparation systemsand improving the efficiency of N fertilizer.

On a national basis, the yield-gap is estimated at 1.5 MT/ha and bridging the yield-gap over the 150,000 ha in production will result in an annual production increase ofover 200,000 MT (table 11). This represents a 33% increase in production and anannual farmgate value of approximately US $33 million.

3.A.3.4.3. Tropical Central America - Panama, Costa Rica, Nicaragua andGuatemala

Traditionally, much of the rice production in Central America came from the uplandsystem, often practiced by small growers on hillsides using a slash/burn, shiftingcultivation system. In this situation rice production was a major contributor to soilerosion and natural resource destruction. The trend toward free trade has exposednoncompetitive production to market forces, resulting in a decline in uplandproduction. This has been particularly true in Panama and Guatemala during the1990s as illustrated by the information in tables 3.24 and 3.25. Costa Ricaexperienced a similar trend during the late 1980s but the area planted to ricestabilized during the 1990s as illustrated in table 3.26. Rice production in Nicaragua,similar to the entire agricultural sector, has been in a phase of recuperation followingthe civil disturbances of the 1970s and 1980s and the annual growth in area plantedto rice during the 1990s surpassed 6% (table 3.27).

Most of the Central American countries have adequate land and water resources tosupport sufficient area of irrigated rice to satisfy national demands. Currentproduction is a mixture of rainfed lowland, upland and irrigated. All four countries ofconcern recognize the potential for irrigated rice and have organized growerassociations to strengthen research and technology transfer in the irrigated sector.The transformation to irrigated rice will be a gradual process and increasing the yieldin the irrigated sector will accelerate the process. Current yields under irrigation arelow, seldom exceeding 4 MT/ha, even though improved varieties for irrigatedproduction are widely used. The principal constraint to increased yields isinadequate crop management, due primarily to limited experience with irrigated rice.

175

The yield-gap is most apparent in the favored rainfed and irrigated ecologies. It isestimated that these two systems comprise about 50% of the total rice area in eachcountry. The potential yield in the two favored systems is 5 - 6 MT/ha, resulting in ayield-gap of 2 - 3.5 MT/ha (table 3.28). Bridging the yield-gap in the favoredproduction systems would increase overall annual rice production in the fourcountries, from the current level of 618,000 MT to 1,157,000 MT. Additionally, theincreased output from the more productive systems will stabilize national suppliesand reduce the need to produce rice in the environmentally sensitive, unfavoredupland system.

• Cuba

Cuba, along with the Dominican Republic and Guyana, is one of the Caribbean'smain rice producers. However, production has been stagnant for over a decade.Rice production in Cuba is limited by the availability of water. Competition for waterfor human use and cultivation of more lucrative and less water demanding crops willfurther restrict the production of rice. Production data indicates that rice productionhas decline from and annual production of over 530,000 MT in the early 1980s toless than 390,000 MT in the 1999 harvest. Information in table 3.29 showsproduction trends during the last decade. Overall yields remain low at 2.4 MT/ha andeven in the areas with adequate water yields, seldom surpass 4.0 MT/ha.

Approximately 160,000 ha are cultivated annually in Cuba but it is estimated onlyabout 30% of the area or 50,000 ha have suitable water to support high yields. Thepotential yield on the 50,000 ha of favored areas is estimated at 6 MT/ha, resulting ina yield-gap of 3.6 MT/ha. Stimulating yield through improved crop management inthe areas highly favorable for irrigated rice production will permit better use of scareresources in Cuba, including water. While part of the problem is related to scarceforeign exchange, there is ample room for more precise and efficient applications ofpesticides and fertilizer.

3.A.3.4.3.1. Summary of yield-gap in Tropical South and Central America andCuba

Total annual rice production of the seven countries in the tropical zone isapproximately 3.1 million MT. The average yield from all production systems is 4MT/ha. However, yields in much of Central American are below 3 MT/ha. In themore productive area comprised of irrigated and favored rainfed, the yield potentialis estimated at 5 - 6 MT/ha. Overall, it is estimated that the yield-gap in the tropicalregion is 1.2 MT/ha (table 3.30). Bridging the yield-gap would increase totalproduction by more than 900,000 MT, annually. This represents a 30% increase intotal rice output and has a commercial value at the farmgate level of approximatelyUS $135 million. In addition to the economic gains derived by bridging the yield-gap,there are positive environmental impacts, especially in Central America, fromfocusing production in areas more appropriate for rice production.

176

Impacts from Bridging the Yield-Gap

The yield-gap is apparent in all countries analyzed; however, the size of the yield-gap varies across regions, between countries and within production zones asillustrated in the summary information presented in table 19. The yield potential forirrigated rice is higher in the temperate region of the Southern Cone due to morefavored climatic conditions. In addition to higher yield potential, the Southern Conehas ample resources for expanding irrigated rice production. Bridging the yield-gapwould not only increase productivity from current area in production but also providethe economic incentive for further expansion.

In the tropical zone, the yield potential for irrigated rice is lower but the yield-gap isapproximately the same when compared to the temperate region (table 3.30). This isdue to lower current yields in tropical South and Central America. The estimatedyield-gap for this region takes into account only areas capable of supporting highproductive rice, i.e., irrigated and favored rainfed systems.

The increase in production feasible by bridging the yield-gap in the 12 countriesanalyzed is estimated at 2.7 million MT (table 3.31). This is equivalent to a 27%increase in production and will contribute annually more than US $ 400 million to thegross income of rice growers. The technology for bridging the yield-gap is availablebut must be introduced, modified to suit local conditions and more importantlyextended to growers. Technology transfer is the key ingredient to bridging the yield-gap and the focus on grower associations provides the means for transferringtechnology in an economical and sustainable manner.

Table 3.13. Evolution of rice production in Latin America and the Caribbeanduring the last four decades

PERIOD AREA YIELD PRODUCTIONannual growth rate (%)

1961-69 3.8 0.6 3.2

1970-79 2.9 0.7 3.6

1980-89 -0.5 3.0 2.5

1990-99 -0.7 3.3 2.6_______Source: FAOSTAT DATABASE. 1999.

177

Table 3.14. Brazil, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99

PARAMETER ANNUAL % GROWTHRATE FOR 1990s

MEANS FOR 1997-99

YIELD 4.25 2.78 MT/ha

AREA -2.86 3,388,600 ha

PRODUCTION 1.26 9,418,600 MT

Table 3.15 Regional production in Rio Grande do Sul, Brazil and estimates ofyield gap in various production zones.

REGION AREA ACTUAL POTENTIAL POTENTIAL INCREASE

YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION

000HA

MT/HA

000 MT MT/HA

OOO MT MT/HA

000 MT

FronteriaOeste

196 5.7 1126 7.0 1372 1.3 254

Litoral Sul 156 5.5 858 7.0 1092 1.5 234

Campanha 128 5.2 664 6.5 832 1.3 168

Depression Central

117 5.2 605 6.0 702 0.8 97

PlanicieCosteraInterna

90 4.7 426 6.0 540 1.3 114

PlanicieCosteraExterna

92 4.3 396 5.5 506 1.2 110

TOTAL/MEANSFOR STATE

779 5.2 4075 6.5 5044 1.3 969

______________Source: Data for actual production provided by IRGA Safra, 1995/96, potential yield and production information extracted fromFLAR Consultancy Report of Edward Pulver and Peter Jennings, June 1998.

178

Table 3.16. Current and potential yield and production of irrigated rice in thevarious regions of Santa Catarina, Brazil.

REGION AREA ACTUAL POTENTIAL POTENTIALINCREASE

YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION

000 HA MT/HA 000 MT MT/HA OOO MT MT/HA 000 MT

Baxixo eMedio Valedo Itajai

13 5.6 73 6.5 85 0.9 12

LitoralNorte

25 5.8 145 7.0 175 1.2 30

Alto Valedo Itajai

9 7.0 63 7.5 68 0.5 5

LitoralCentro

3 4.7 14 6.0 18 1.3 4

Litoral Sul 16 5.4 86 6.5 104 1.1 18

Sul doEstado

50 5.6 280 7.0 350 1.4 70

TOTAL/MEANS forState

116 5.7 661 6.9 800 1.2 139

______________Source: Current production and yield information obtained from "Diagnostico da Estrutura do Produca de Arroz Irrigado enSanta Catarina 1997. XXXII. Reuniao da Cultura do Arroz Irrigado. EPAGRI. Potential yield and production obtained fromFLAR - Consultancy Report "Current and potential rice production in Santa Catarina". Edward Pulver. 1999.

179

Table 3.17. Argentina, annual growth rates in yield, area and production ofrice during the 1990s and mean yield, area and total production for the three-year period of 1997-99


MEANS FOR 1997-99


AREA 11.6 217,000 ha

PRODUCTION 14.4 1,085,200 MT

Table 3.18. Uruguay, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99.


MEANS FOR 1997-99


AREA 8.76 177,696 ha

PRODUCTION 11.37 1,078,974 MT/ha

Table 3.19. Bolivia, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99.


MEANS FOR 1997-99


AREA 1.32 126,000 ha

PRODUCTION 2.28 254,500 MT

180

Table 3.20 Chile, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99


MEANS FOR 1997-99

YIELD -0.17 4.0 MT/ha

AREA -1.49 25,200 ha

PRODUCTION -1.66 108,000 MT

Table 3.21. Current and potential yield and production of irrigated rice inFLAR-member countries located in the temperate regionCOUNTRY AREA ACTUAL POTENTIAL POTENTIAL

INCREASE000 HA YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION

MT/HA 000 MT MT/HA OOO MT MT/HA 000 MT

Brazil

- RS 779 5.2 4075 6.5 5044 1.3 969

- SC 116 5.7 660 6.9 797 1.2 137

Argentina 217 5.0 1085 7.0 1519 2.0 427

Bolivia1 126 2.0 252 5.01 2821 3.01 301

Chile 27 4.0 108 6.0 162 2.0 54

Uruguay 164 5.7 935 7.0 1148 1.3 213

TOTAL/MEANS FORTEMP.REGION

1429 5.0 7115 6.5 8952 1.3 1830

__________1 All rice production in Bolivia is upland, program promotes the conversion of 10,000 ha to irrigated rice with an average yieldof 5.0 MT/ha during the three-year project proposal.Source: Information from Brazil extracted from tables 2 and 3 of main text and data for other countries extracted fromFAOSTAT, 1999.

181

Table 3.22. Current and potential yield and production in the major irrigatedrice growing regions of Colombia.


000 HA YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION


Centro 105 6.3 662 7.5 788 1.2 126

Llanos 76 4.9 372 5.3 403 0.4 31

Bajo Cauca 41 4.2 172 5.5 226 1.3 54

Costa Norte 28 5.2 146 6.0 168 0.8 22

Santanders 22 4.9 108 5.5 121 0.6 13

TOTAL/MEANS FORCOLOMBIA

272 5.4 1460 6.3 1706 0.9 246

_____________Source: Current data for area, yield and production obtained from "Arroz en Colombia", 1997. FEDEARROZ. Division deInvestigaciones Economicas. Information on yield potentials extracted from FLAR-FEDEARROZ 1996. Edward Pulver andPeter Jennings.

Table 3.23. Current and potential yields and production of irrigated rice in thetwo main production zones of Venezuela.


YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION000 HA MT/HA 000 MT MT/HA OOO MT MT/HA 000 MT

LlanosCentrales

62 3.8 236 5.0 310 1.2 74

LlanosOccidentales

88 4.9 429 6.5 572 1.6 143

TOTAL FORVENEZUELA

150 4.4 665 5.9 882 5.9 217

_______________Source: Data on current yield and production extracted from APROSCELLO, 1999. Information on potential yield andproduction extracted from FLAR-Fundarroz, 1997 Consultancy Report "El Papel de Fundarroz in la industria arrocera enVenezuela". Edward Pulver and Peter Jennings.

182

Table 3.24. Panama, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99


MEANS FOR 1997-99


AREA -5.92 57,000 ha

PRODUCTION -5.45 143,000 MT

Table 3.25. Guatemala, annual growth rates in yield, area and production ofrice during the 1990s and mean yield, area and total production for the three-year period of 1997-99


MEANS FOR 1997-99


AREA -2.35 13,000 ha

PRODUCTION -1.80 38,000 MT/ha

Table 3.26. Costa Rica, annual growth rates in yield, area and production ofrice during the 1990s and mean yield, area and total production for the three-year period of 1997-99


MEANS FOR 1997-99

YIELD 0 3.5 MT/ha

AREA 1.35 65,000 ha

PRODUCTION 1.35 228,000 MT/ha

183

Table 3.27. Nicaragua, annual growth rates in yield, area and production ofrice during the 1990s and mean yield, area and total production for the three-year period of 1997-99


MEANS FOR 1997-99


AREA 6.07 72,000 ha


Table 3.28. Current and potential yield and production of irrigated rice ofFLAR-member countries in the tropical region (excluding Colombia andVenezuela)

Country Area Actual Potential Potential Increase

000 Ha Yield Production Yield Production Yield Production


Costa Rica1 65 3.5 228 6.01 3091 2.51 811

Guatemala1 13 2.9 38 5.01 521 2.11 141

Nicaragua1 72 2.9 209 5.01 2841 2.11 751

Panama1 57 2.5 143 6.01 2421 3.51 991

Cuba2 160 2.4 384 6.02 5642 3.6 1802

Total ForTropicalRegion

367 2.7 1002 5.8 1451 1.2 449

_________1 Countries have a combination of upland and irrigated rice, program will focus on improving production on 50% of currentarea, where water does not limit productivity.2 Water deficits limit improvements for entire 160,000 ha under cultivation, program will focus on improving yield to 6 MT/ha on50,000 ha that have adequate water resources.Source: Current production data are three-year means of 1997-99, extracted from FAOSTAT Database, 1999. Yieldpotentials estimated by FLAR personnel, October, 1999.

184

Table 3.29. Cuba, annual growth rates in yield, area and production of riceduring the 1990s and mean yield, area and total production for the three-yearperiod of 1997-99


MEANS FOR 1997-99


AREA 1.14 160,000 ha


Table 3.30. Current and potential yield and production of irrigated rice ofFLAR-member countries located in the tropical region

COUNTRY AREA ACTUAL POTENTIAL POTENTIALINCREASE

000HA

YIELD PROD YIELD PROD YIELD PROD

MT/HA

000 MT MT/HA OOO MT MT/HA 000 MT

Colombia 272 5.4 1460 6.3 1706 0.9 246

Venezuela 150 4.4 665 5.9 882 1.5 217

Costa Rica1 65 3.5 228 6.01 3091 2.51 811

Guatemala1 13 2.9 38 5.01 521 2.11 141

Nicaragua1 72 2.9 143 5.01 2841 2.11 751

Panama1 57 2.5 143 6.01 2421 3.51 991

Cuba2 160 2.4 384 6.02 5642 3.62 1802TOTAL/MEANSFOR TROPICALREGION

789 4.0 3127 6.0 4039 1.2 912

____________1 Countries have a combination of upland and irrigated rice, program will focus on improving production on 50% of currentarea, where water does not limit productivity.2 Water deficits limit improvements for entire 160,000 ha under cultivation, program will focus on improving yield to 6 MT/ha on50,000 ha that have adequate water resources.Source: Information for Colombia extracted from information provided by FEDEARROZ, information for Venezuela wasprovided by APROSCELLO, and information from other countries was extracted from FAOSTAT, 1999.

185

Table 3.31. Overview of current and potential yield and production of irrigatedrice in FLAR-member countries in both major production regions

REGION/COUNTRY

AREA ACTUAL POTENTIAL POTENTIALINCREASE

000 HA YIELD PRODUCTION YIELD PRODUCTION YIELD PRODUCTION


TEMPERATE

- Brazil

- RS 779 5.2 4075 6.5 5044 1.3 969

- SC 116 5.7 660 6.9 797 1.2 137

- Argentina 217 5.0 1085 7.0 1519 2.0 427

- Bolivia1 126 2.0 252 5.01 2821 3.01 301

- Chile 27 4.0 108 6.0 162 2.0 54

- Uruguay 164 5.7 935 7.0 1148 1.3 213

REGIONTOTAL/MEANS

1429 5.0 7115 6.5 8952 1.3 1830

TROPICAL

- Colombia 272 5.4 1460 6.3 1706 0.9 246

- Venezuela 150 4.4 665 5.9 882 1.5 217

- Costa Rica2

65 3.5 228 6.02 3092 2.52 812

- Guatemala2 13 2.9 38 5.02 522 2.12 142

- Nicaragua2 72 2.9 209 5.02 2842 2.12 752

- Panama2 57 2.5 143 6.02 2422 3.52 992

- Cuba3 160 2.4 384 6.03 5643 3.63 1803

REGIONTOTAL/MEANS

789 4.0 3127 6.0 4039 1.2 912

TOTAL/MEANS FORFLAR-MEMBERS

2218 4.6 10248 6.3 12991 1.2 2742

________________1 All rice production in Bolivia is upland, program promotes the conversion of 10,000 ha to irrigated rice with an average yieldof 5.0 MT/ha during the three-year project proposal. 2 Countries have a combination of upland and irrigated rice, program willfocus on improving production on 50% of current area, where water does not limit productivity. 3 Water deficits limitimprovements for entire 160,000 ha under cultivation, program will focus on improving yield to 6 MT/ha on 50,000 ha that haveadequate water resources.

186

OUTPUT 3. ENHANCING REGIONAL RICE RESEARCH CAPACITIES ANDPRIORITIZING NEEDS WITH EMPHASIS ON THE SMALL FARMERS

3.B. Rice Economics

3.B.1. Update estimates of adoption in LAC of new rice varieties by area andproduction, 1970-99 (with IFPRI and with Impact Unit of CIAT)

Luis R. Sanint, CIATPhil Pardey, IFPRI

Eduardo Castelo Magalhaes, IFPRI/ EMBRAPAMarco A. Oliveira, FLAR

Milestones: data on varietal use 1970-99, main producing countries in LAC: Brazil,Argentina, Uruguay, Colombia, Peru, Ecuador, Venezuela.

Budget: US$20,000 per year from CIAT; US$60,000 from IFPRI.

Trips were made and researchers were contacted to gather up-to-date data on theevolution of varietal use. Collaborators and countries visited are:

Temperate region:Brazil (EMBRAPA, IRGA, EPAGRI)Argentina (CIALA, INTA)Uruguay (INIA)

Tropical region:Colombia (FEDEARROZ)Ecuador (INIAP, FENARROZ)Peru (Seed Chamber)Venezuela (FUNDARROZ)

3.B.2. Coordinate meetings and activities of the Latin American Network ofRice Economists

Milestones:Meeting of the network: not accomplished due to lack of fundingInterchange of information: not achieved, as network is not a top priority of FLAR

3.B.3 Participate in study of IPRs in rice in LAC, with IFPRILuis R. Sanint, rice economist, CIAT

Eran Binenbaum, U. of California, Berkeley

Milestones: articles published. Not done. Student postponed research until nextyear.

187

3.B.4. Monitor world rice marketsLuis R. Sanint, rice economist, CIAT

As a service to FLAR members, FLAR publishes twice a year a summary of worldrice conditions in its Foro Arrocero.

3.B.5. Additional activities not included in plan.Luis R. Sanint, rice economist CIAT

Myriam C. Duque, statistical consultant, CIATNestor Gutierrez, Director Economic Research, FEDEARROZ - Colombia

Miguel Diago, Technical Director, FEDEARROZ - Colombia

Milestones: study on the adoption of new rice technologies in Colombia

Latin America, with 24 million tons of paddy production in 1999, contributed about4% of the rice globally produced. In this region of the world, the technologyknown as the “Green Revolution” was implemented since its inception in the1960’s, and it was precisely this technological change that resulted in a 300%increase in production over the past four decades. The growth in production wasmainly due to radical increases in productivity (Sanint, 2000).

Colombia has been a leader in the adoption of the new technologies, representedto date in the appearance of high-yielding dwarf varieties, which have made theexpansion of this crop possible. In 1999 Colombia produced 2.3 million tons ofpaddy rice (Fedearroz, 2000), placing the country in second place for productionin Latin America, after Brazil, which accounts for half the regional rice output.

In Colombia rice occupies first place in economic value among the short-cyclecrops; furthermore, it is the primary source of calories and proteins for the 20%group with lower income among Colombians. Two crops can be grown in mostregions, depending, mostly, on water availability. A comparison of results from theFirst and the Second National Census of Rice Growers, which were carried outduring 1988 and 1999, respectively, reveals, in the last decade, the number ofproducers increased by 60%, while the area cultivated increased only by 30%.(Fedearroz, 1990; Fedearroz, 2000). This implies a reduction in the average areaof the rice plots (from 14 has in 1988 to 10 has in 1999) and an increase inproductivity (from 4.4 tons/ha in 1988 to 4.8 tons/ha in 1999). By 1999 cultivatedarea reached 493,237 has (Fedearroz, 2000).

There are two distinct production systems in the country: mechanized and manual–or traditional-- upland rice. Mechanized rice represents 95% of the rice area and98% of production. However, traditional upland rice has a social value in specificregions of the country, as almost half of rice producers in the country (47% of atotal of 28,128 rice farmers) are found in this system. Average yields are only 1.1tons/ha. That compares with 4.8 tons/ha in mechanized rice, which, in turn, isdivided in irrigated (with 63% of the rice area and a yield of 5.4 tons/ha) andrainfed lowlands (with a share of 32% of the area and a yield of 4.2 tons/ha). The

188

latter differs from the former with respect to its dependence on rainfall, which isgenerally concentrated in a few months of the year, thereby determining the verymarked cyclical nature of production in this cropping system. Of the area plantedto mechanized rice, 67% corresponds to the irrigated rice system and 33% to therainfed lowlands system. In terms of production, the irrigated rice systemcontributes 72% of mechanized rice output, in contrast to 28% for the rainfedlowlands rice system, which points at marked differences in productivity.

Geographically speaking, rice production in Colombia is divided into fourproduction zones (see next section on materials and methods, for details). Each ofthese regions has different agroecological and edapho-climatic properties, whichcharacterize and differentiate them (Fedearroz 1998).

Another important aspect that should be highlighted in relation to thecharacteristics of the rice growers is related to land tenure; 49% of the ricegrowers are owners and 51%, renters. This means that production costs inColombia must take into account the leasing of the land within the variable costsof the crop.

The objectives of the paper are to examine the role of several socioeconomicvariables in the determination of yield gaps and to determine if there is evidenceof declining yields in rice in Colombia.

3.B.5.1. Materials and Methods

3..B.5.1.1. Definitions

Zone: each one of the regions into which the country has been divided and whichconstitute very different agroclimatic environments.

ZonePart of the

Country ProvincesCauca RiverWatershed

Southwest Cauca

Center Center Tolima, Huila,Valle del Cauca,Cundinamarca

North (Atlantic) Coast North Caribbean regionEastern Plains East Meta

Harvest A: as many areas harvest twice per year, this category refers to theproduction of the rice plots planted during the first semester. It is the largest of theyear by area and production, accounting by some 60% of the yearly total.

189

Harvest B: covers the rest of the production, the one obtained from plantingsduring the second calendar semester. A good part of this harvest takes placeduring the first months of the following calendar year.

Size: based on the area planted to rice, three categories were generated:

Categories Defined as:Small Less than 3 haIntermediate From 3-10 ha (inclusive)Large More than 10 ha

Rice cultivation system (mechanized): this classification is based on the use andavailability of water. Basically there are two systems: Irrigated and Rainfedlowlands. The latter depends on rainfall; but in terms of management, it uses thesame practices and technology as the former.

Land Tenure: this category is based on the condition of ownership (or not) of thelot where the rice crop is grown. Two levels were considered: “renters” and“others,” which combines the remaining possibilities (owners, untitled farmers,etc.).

Cells: they refer to categories resulting from the interactions of Zone, System,Land Tenure and Size.

Percentiles: they were generated for the producers of the sample as a function ofthe results obtained in terms of the variable of interest —yields. For this purposethe n values were organized in ascending order, giving rise to the set (X1, X2, X3,…, Xn) and the tst percentile was defined as the value p = t/100

np = j + g; j = the whole part of the yieldg = the decimal part of the yield

the t-percentile is defined as the weighted average of Xj and Xj+1, values that flankthe value associated with the percentile of interest.

Y = (1 - g) Xj + g Xj+1

For purposes of the analysis, the variable "Type of Producer" was formed forthose with yields below percentile 10 and above percentile 90; in other words, the10% of the producers with the lowest yields and the 10% of the producers with thehighest yields throughout the 1991-98 period.

3.B.5.1.2. Data sources

Censi: The I and II National Rice Growers Census, with the 1988 and 1999 (bothA-B harvests), conducted by FEDEARROZ.

190

National rice growers survey: from the1988 Census, the stratification of thecountry was done in terms of:

Size (three levels)Land Tenure (two types)Cultivation System (two classes)Zone (four regions)

This resulted in 48 cells of interest.

An optimal sample size of 180 producers was randomly selected at a 97%confidence level of confidence (based on yield variability), and this was appliedproportionately among the cells defined by the strata. For each farmer selected,two substitutes with similar characteristics were identified in the event that theinterview could not be conducted with the first one so as to maintain theexpansion factors of each farmer in the sample valid.

The sample had a trial period in 1989 and 1990 and has been formally appliedsince the 1991 Harvest A until now. The paper includes information up to the 1998Harvest A. Since two harvests are evaluated each year, there are 15 observationsper variable in the 1991-98 sample.

Experimental data: the information was obtained from FEDEARROZ onexperimental data in demonstration plots and regional trials of the varietyFedearroz-50.

3.B.5.1.3. The analysis

The principal part of the analysis of the data relates to crop management aspectsin the sample. The 99 Census and the experimental data on Fedearroz-50 aretaken as points of reference. Fedearroz-50 is the most recent variety released andin the process of being adopted in the country during the last semesters in thetime span of the sample.

Census 99. Contingency tables were developed in order to test the hypothesis ofindependence between the variables Type of Producer and each of the following,using the χ2 statistics:

municipality (or “township”)agelevel of schoolingyears growing riceyears on the farm

These tests were run for each one of the cells (Zone-System–Land Tenure-Size).

191

The analyses were repeated for:

Ibagué-Espinal-SaldañaIbagué-Espinal-Saldaña-Purificación-GuamoCauca ProvinceValle del Cauca Province

National rice growers survey

Descriptive study. For each possible combination among Zone, System, LandTenure, Size and Period, the average yields (t/ha) were calculated using the SASMEANS procedure, weighted by the number of hectares represented by eachobservation.

In addition, maximum/minimum variances, coefficients of variability and totalvariance were calculated.

Trends. In order to detect trends in yield levels, two regression models wereadjusted, using the SAS REG procedure in each one of the previously definedcells, omitting Size and Land Tenure.

The first model was a simple linear regression: y = a + b X + ε

and the second was the exponential defined by the expression y = A eb*time.

The study was repeated separately for Zone and System within the variables thathad sufficient data over time in order to detect trends within them.

The best and worst producers: the purpose was to determine whether theproducers labeled as the best (yields above percentile 90) or the worst (yieldsbelow percentile 10) remained in the same classification consistently over time. Aquantitative analysis was run, generating the categorical variables "consistentlylow," "consistently intermediate," "consistently high" and "not consistent"; and therespective table of frequencies was constructed.

Experimental data: the confidence intervals were calculated for the yields ofFedearroz-50 to see whether the yields reported by the 99 Census should beincluded.

3.B.5.2. Results and Discussion

3.B.5.2.1. Experimental Results versus National Average Yields by Zone

Traditionally the technological gap is measured by the difference in productivityobtained at the experimental level and the lower yields that the producers obtainin their farms. In order to measure this gap, the results obtained by FEDEARROZresearchers in the commercial and semicommercial trials of their most recently

192

developed variety Fedearroz-50, which was turned over to the producers in thesecond semester of 1999, were used. To obtain the results at the producer level,data on the yields obtained in the National Census of Rice Growers for the sameperiod were used. When the 96 experimental results in the four nationalproduction zones were grouped and compared with homologous results—theaverages obtained by the Census—some very interesting results were found.

First of all, the two zones where the above average experimental data weregreater than the observed in the Census (Cauca River Watershed and NorthCoast), represent only 20% of the national production. In the Central Zone and inthe Plains, which are the two main rice-producing centers, producers achievedyields that were either the same (Central Zone) or significantly greater (in thePlains) than those of the experimental plots: 5360 kg vs. 4773 Kgs. These resultsgive a very good idea of the existing gap between the two data sets beingcompared. The gap is somewhat different from what is usually reported in themajority of the countries around the world, where the experimental results aregreater than those obtained directly by top farmers under actual productionconditions.

The question arises as to how this type of gap, ”contrary to the traditional concept”can be generated. This is where we have to recur to the particularities of thegeneration and diffusion of technology in Colombia and, to the extent possible, thenature of the Colombian rice growers.

As of 1990, when the responsibility for biological and economic research wasmostly transferred to the private sector (FEDEARROZ), some of the strategiesthat are being implemented today were reformulated then, first with respect tosetting research priorities and second, for selecting the experimental areas. In theformer case, FEDEARROZ consulted with the scientific community to better focusthe biological research and prioritize the sector's needs from this perspective. Atthis moment, the most important factor for allocating the distribution of the fundsfor research is the direct consultation with producers, most of whom expressedtheir interest through their direct involvement in the development of new varieties.

In this process of reorienting the research, it was also necessary to define clearlythe sites where the research would be carried out. Thus, in addition to carryingout the primary and basic steps of all the processes on the experimental farmssituated in all the rice-producing zones of the country, the majority of the trials anddemonstration plots were transferred directly to the producers' farms. Some of theexperimental centers are located in areas where there is high disease pressure(“hot spots”) in order to select the materials that are most resistant to pests anddiseases. It should be no surprise that rice cultivated in those plots does notachieve the best expression for yield potential. With this strategy, extraordinaryresults were obtained at the moment of disseminating the technology. Theproducers themselves acted as multipliers; thus the time required for this diffusionwas surprisingly short. In the case of the variety Fedearroz-50, it was a matter of

193

only a few months before it was planted throughout the country, its dispersiondepending only upon the availability of certified seed.

Another reason why the technology was adopted rapidly by the great majority ofthe producers in Colombia lies in the simplicity of the technological package,which is easy to understand and adopt. An attempt will now be made to explainjust how homogeneous is the use of this technological package within the differentrice-growing regions by focusing on the relative gaps among groups of farmersand on the factors that might explain them.

3.B.5.2.2. Yield Gap Among Producers by Zone

Based on the experimental results and the productivity at the producer levelpresented in table 3.32, it is evident that there are significant differences in theaverage yields obtained in the different rice-producing zones of the country. Thisis hardly surprising given the differences in their geographic localization, types ofsoils, and availability of water, above all. At first glance, there is a wide gapbetween manual and mechanized rice (1.1 tons/ha versus 4.8 tons/ha). Obviously,edapho-climatic conditions explain a big share of that gap. To draw meaningfulinsights into the variables that determine yield gaps from the perspective of cropmanagement, it is important to isolate those external factors and approach theissue at the level of homogeneous groups of farmers. Therefore, the analysisfocuses on mechanized rice, both irrigated and rainfed lowlands. It should benoted that mechanized systems use very similar technologies, including thevarieties and the rest of the production inputs, while traditional upland rice isdrastically different in almost all its technological components. The paper looks,first, at gaps in the national aggregates and gradually narrows the focus tosuccessively concentrate on more homogeneous sets of farmers, in an effort toelicit the effects of a group of selected variables in the observed yield gaps. Itwould be interesting to learn how homogeneous the production yields are withineach region. It is known that there are differences among regions, but it would beof great help to learn whether there are large differences among the producers ofeach region.

Using the results of the 1999 Census, which covered most of the 28,000producers, the rice growers were grouped within each zone according to theirproduction system in order to explain the degree of variance better. In order tolearn whether the degree of dispersion within each group of producers was verylarge or acceptable, simple coefficients of variability were used that did not exceed25% as the critical level for distinguishing between these two categories.

Table 3.33 gives the average productivity for each one of the regions byproduction system. The coefficient of variability was higher than 25% only in therainfed lowlands rice system in two regions—the Cauca River Watershed and theNorth Coast.

194

Another important aspect to be highlighted in table 3.33 is that the producers ofirrigated rice, who account for 71% of the national production, do no present ahigh degree of variability. Rainfed lowlands rice production, of course, has agreater risk factor given that it depends exclusively upon rainfall, in spite of thesimilarity of the technological package used for irrigated rice. Within the rainfedlowlands rice system, the region of the Eastern Plains, which accounts for 90% ofthis production, there was no level of variance greater than 25%. Thus it can beconcluded that within the regions, the great majority of the rice production,independent of the production system, does not present significant variances withrespect to the average level of productivity. Thus the production gap amongproducers by region is a problem specific to the rainfed lowlands rice productionsystem in two production zones.

3.B.5.2.3. Behavior of the Gap in the Central Zone

In order to understand better the behavior patterns of the dispersion of productivitywithin the region, the Central Zone1, the most important irrigated rice productionzone of the country, was selected. The purpose was to break down of thebehavior of the best and the worst producers so as to identify some variables oftraditional analyses to explain this variability.

In order to do this analysis, the technique of extreme deciles was used. Thisconsists in dividing the total production into ten groups. For the analysis, the twodistribution extremes were compared, which means working with the 20% of thepopulations that contrast the most.

According to the researchers' criteria, the type of land tenure and the size of therice plot could be two variables that help to explain this difference. Thus for theCentral Zone, it was decided to group producers by ownership (owners andrenters) and by size. The results show differences of almost 50% in yieldsbetween the best and the worst producers. The first reaction to these resultswould be how could such a large difference in yields between these twocategories occur when the total for the region does not present such a greatvariability as shown previously? The response to this question is that 80% of theproducers have near average yields; in other words, the distribution of the yieldsapproaches a normal distribution. The analysis by farm size reveals that there isno significant gap among owners or renters; thus it is possible to begin to excludethis variable as being responsible for the differences. It should be noted that in thisregion of the country for the year 1999, Semester A, the Census revealed that65% of the producers corresponded to the category of renters. Table 3.34B showsthat the same pattern of distribution for the gap is maintained for both the rentersand owners, ruling ownership as a significant variable as well.

1 This is a highly homogeneous region that includes the four provinces of Tolima, Huila, Valle andCundinamarca)

195

In order to explore this issue further and try to find logical explanations tocharacterize the best and worst rice producers in the Central Zone, it was decidedto explore another group of variables that would explain the gaps observed, atleast partially. The 1999 Census quantified variables of a social nature, such asthe farmer's age, level of schooling, years on the farm and years as a rice grower.The variable "township" (the smallest political-administrative division within theprovinces), which corresponds to the geographic characterization, was alsoincluded. To test the hypothesis of independence for each of these five variableswith the good or bad growers, contingency tables were used with the Chi squarestatistical test at the 5% level.

Among all the social and geographic variables analyzed, only one shows astatistically significant relationship with yield: “township”. Thus the hypothesis ofindependence between the productivity of the producers and the socioeconomicvariables studied could not be rejected. That is, the grower's performance (goodor bad) within a given township does not depend upon the educational level, ageor experience as a farmer or as a rice grower. Rather, it may be a function of othervariables not included in the analysis, like individual talent or other randomfactors.

3.B.5.2.4. Gap at the Township Level

Based upon this last hypothesis, the level of analysis was refined, taking thetownship as the focal point for the new analysis, to achieve a very high level ofhomogeneity within the sample. Within the Central Zone the three most importanttownships were selected and grouped. In order to obtain a sufficient number ofproducers based on this new grouping, the previous analysis was done for theentire Central Zone.

Results of this new level of analysis coincide with the data presented in tables3.34A and 3.34B. Nevertheless, it should be noted that the size of the gapbetween the best and the worst producers decreased to 30%. This last effect maybe due to the selection that was made of the townships. That is to say, that up tothis point, the variable township is the only one that explains, at least partially, thegap that exists between the two types of rice growers selected.

As for the previous case, rice plot size and type of ownership did not explainchanges in the behavior of the gap. The analyses of tests for independenceamong the social and geographic variables and the classification of the growersas good or bad were repeated. No different associations from that found for thevariable township were discovered.

In this state of analysis, there is room for asking what other type of variables couldexplain the gap. One quick answer can be the individual skills and talent of theproducer. There can also be, of course, other random variables characteristic ofthis activity—such as climate, timing in the execution of cultural practices,opportunity in credit disbursements, availability of labor, quality of inputs, etc. In

196

the case of rice in the rainfed lowlands, the timing and amount of rains are crucial.For the country as a whole, the precise application of management practices canbe severely distressed by the country's social unrest, particularly in the ruralareas.

3.B.5.2.5. Randomness in Producers' Behavior

There are good growers and poor growers; the skills of a given farmer couldexplain a good portion of the gap, when compared to other farmers. In order toexplore this issue more in depth, the behavior of a group of rice growers from theCentral Zone, selected within the National Rice Growers Survey, was studied indetail during the period 1991-1998. This period includes 15 cropping seasons.The purpose was to determine whether during the 15 semesters studied, theproducers categorized, according to their performance, among the worst 10% oramong s the best 10%, maintained their position throughout the period. If so, thiswould mean that there was one group whose behavior contributes to explain thegap in a systematic way, decreasing the participation of other random factors. Itcould be said, for example, that the edapho-climatic characteristics of themicroregion or their particular skills could result in their always being classifiedwithin the group of the worst or within the group of the best producers. If it werefound that the same grower was some semesters in the group of the worst, othertimes in the best and others in the average group, we could conclude that thereare other random factors that determine productivity gaps.

As a basis for these analyses the information from the National Rice GrowersSurvey was used. In the case of the Central Zone, 64 farms were monitored,starting from the first semester of 1991 through the first semester of 1998, for atotal of 15 continuous semesters for each rice grower in the sample.

Table 3.35 shows the behavior of the producers surveyed, categorized, by rangesof performance, in 3 groups based on the yields obtained throughout the years:among the worst 10%, the 80% intermediate and the best 10%. The results arevery interesting: first, it can be seen that of the 26 renters in the irrigated ricesystem, only 2 were consistently found among the 10% with the lowest yields, 8producers maintained their position in the intermediate group, and the remaining16 moved within the three categories. These results show that only 2 out of the 26producers maintained their position in the category of the worst producers overthe 15 semesters; the remaining 24 moved randomly within the 3 categories.

In order to corroborate this analysis, the group of 39 land owners was submitted tothis procedure (table 3.36). It was found that only 7 of them maintained theirposition in the intermediate group; the remaining 32 moved among the threecategories. In this case, as in the case of renters, the majority of producers are inthe intermediate group most of the time. Once again, this is precisely whatconfirms the low variability that exists in yields within homogeneous rice-growingzones.

197

The partial conclusion of this section is that the producers' classification as a goodor bad performer is variable, changing from semester to semester. When onelooks at what occurs to a given rice grower over 15 semesters, its performance isnot consistently ranked in the same group. The changes observed in performancemost likely respond to variables that he cannot control and that randomly affectthe rice operation. These could be the levels of luminosity per day, temperature,inappropriate technical assistance as well as overall unforeseen factors, such associal unrest -- a variable whose importance should not be neglected in Colombia-- that affect the precision in the execution of specific cultural practices.

Perhaps more important, because of the implications for technology generationand diffusion, is the fact that the results also point at the existence of technologiesthat are very similar, are widely used by most farmers and result in similarperformances when applied in homogeneous locations. In Colombia, there aretwo distinctly different technologies, targeted to different ecosystems: one formechanized rice and another for traditional (or manual) upland rice. While everyfarmer in the mechanized sector uses modern rice varieties, two thirds of thefarmers in manual upland conditions still use traditional varieties (Fedearroz,1990). The evidence corroborates de Janvris observation in the sense that themajority of innovations are likely to be specific to areas owned or controlled byspecific groups (de Janvri, 1973). Even if most current mechanized rice areas inColombia were marginal twenty or thirty years ago, over time they became quiteprosperous as a result of the innovations and the influence on research programsby the groups of organized farmers under FEDEARROZ. At the township level,there does not seem to be farmers with consistently “superior” packages,practices and performance. Which is also equivalent to say that rice technologiesare universal and neutral: they have been readily adopted by all sort of farmersand, within similar environmental conditions, farmers are likely to obtain verysimilar yields. Among locations with highly homogeneous conditions, the gap istherefore, independent of the rice grower's individual skills, the type of technologyand the social variables such as age, experience in farming and in rice productionand level of education.

3.B.5.2.6. Yield Trends in Rice: The 1991-98 Period and Beyond

To determine the productivity trends of the Colombian rice-growing sector, simpleand exponential linear regressions were run using data from the National RiceGrowers Survey of 180 producers; they were classified by zone, productionsystem, type of land tenure and range of area; data were collected during theperiod 1991-1998.

The results of this analysis showed a positive growth trend in the zone of theNorth Coast and in the Central Zone for the producers with larger extensions. Inthe rest of the zones, the behavior of this variable tended to be stable.

For the detailed analysis of the growth zones, it was necessary to break down theproduction data by varieties. It was observed that this growth depends upon the

198

rate of adoption of the higher yielding varieties (Oryzica 1 on the Coast and Roa inthe Central Zone).

What is the future of rice productivity in Colombia? In the year 2001 at least onenew high-yielding variety will be released by FEDEARROZ and in 2002 anothertwo varieties will be placed on the market. Several private seed companies havenew lines in advanced yield trials that are will also be released soon. This ensuresthat the producers will face a growing menu of options. The risk of the country'sproduction being based on a few varieties with a narrow genetic base isdecreasing. Results of this analysis imply that new practices are quickly anduniversally adapted when they prove to be successful in increasing productivity.Therefore, as research in crop management focuses more in the issues ofprecision farming, the new technologies should be quickly adopted by mostfarmers and will contribute to raise the expression of yield potential in the newvarieties. The continuous flow of superior new germplasm and of new cropmanagement techniques should contribute to keep the upward trend in averageyields at the farm level. This is a likely scenario, provided that social unrest isgradually reduced to allow that the rural sector recuperates its leading role in theeconomy and that international trade provides a fair environment for competition.

3.B.5.3. Conclusiones

Colombia has been a leader in the adoption of new rice technologies in thiscontinent over the past three decades. Rice production systems in the country arehighly heterogeneous. The traditional view of yield gaps compares experimentaldata with yields of top producers. In Colombia, this gap is contrary to what hasbeen traditionally hypothesized: top farmers outperform experimental yields. Aplausible explanation is that experiments are conducted in places where pest andother environmental pressures are high and varieties do not express their yieldpotential. To better understand the nature of yield gaps among farmers and thefactors influencing them, the effects of a set of socioeconomic variables wereexplored and robust data sets obtained by FEDEARROZ over the past decadewere used. Yields at the national level exhibit big gaps among groups of farmers.But the gaps are mostly explained by edapho-climatic conditions of each regionand by production systems. In other words, while traditional rice farmers achieve,on average, 1.1 tons/ha, mechanized rice farmers obtain 4.8 tons/ha. Andirrigated rice farmers in the central region obtain 6.1 tons/ha while in other regionsthe average yield for that system is around 4.7 tons/ha. The fact that smallerfarmers exhibit lower yields than larger farmers basically reflects that most smallfarmers are traditional upland rice growers, mainly located in frontier, marginallands. Therefore, it is important to ensure that we do not end up comparing appleswith oranges. Consequently, the sample was grouped in cells formed by theinteraction of four characteristics that define farmer prototypes: regional location,tenure, production system and field size. Contingency tables were developed totest the hypothesis of independence between the variables “Type of Producer”and each of the socioeconomic variables (farmer's age, level of schooling, yearson the farm and years as a rice grower). Tests were run for each one of the cells

199

(Zone-System–Land Tenure-Size) using the χ2 statistics. Among all the social andgeographic variables analyzed, only one shows a statistically significantrelationship with yield: "municipality". Thus the hypothesis of independencebetween the obtained yield by a given producer and the socioeconomic variablesstudied could not be rejected. That is, the grower's performance (good or bad)within a given municipality does not depend upon the schooling level, age orexperience as a farmer or as a rice grower. A quantitative analysis on theperformance of farmers by yield obtained was run, generating the categoricalvariables "consistently low," "consistently intermediate", "consistently high" and"not consistent"; and the respective table of frequencies was constructed. Theconclusion is that the producers' classification as a good or bad performer isvariable, and it differs from semester to semester. The changes observed inperformance most likely respond to variables that cannot be controlled and thatrandomly affect the rice operation. There do not seem to be farmers withconsistently “superior” packages, practices and performance, suggesting that ricetechnologies are neutral. New technologies have been readily adopted by alltypes of farmers. In turn, within highly homogeneous edapho-climatic conditions(at the municipality level), farmers are likely to obtain very similar yields. The gapis independent of the rice grower's individual skills, the type of technology and thesocial variables such as age, experience in farming and in rice production andlevel of education. Such results can only be explained by the fact that the new ricetechnologies are neutral, quite universal, relatively simple, and easily and quicklyadopted. The issue of traditional growers is still pending; they have also benefitedfrom new technologies but to a much lesser degree. Yet, they account for lessthan 2% of the rice output of Colombia. The evidence from studying themechanized rice group confirms that the majority of innovations are likely to bespecific to areas owned or controlled by distinct groups, as is the case of floodedrice (irrigated and rainfed lowlands). Although most rice areas in Colombia weremarginal not many years ago, they became quite prosperous over time as a resultof the innovations and the influence on research programs by the actions ofproperly organized groups of farmers under FEDEARROZ. Finally, the sampledoes not show declining rice yields in Colombia; in most cases, there was not adefinite trend, but when a significant trend in yields was detected among a groupof farmers, it was positive. In the near future, the continuous flow of superior newvarieties will continue. New crop management techniques emphasizing higherprecision in the application of inputs should contribute to favor the expression ofyield potential in the new varieties to ensure an upward trend in average yields atthe farm as well as at the national level.

200

References

De Janvri, Alain. 1973. A Socioeconomic Model of Induced Innovations. QuaterlyJournal of Economics. 87, pp. 410-435.

Fedearroz. División de Investigaciones Económicas. 1990. Primer CensoNacional Arrocero: Cubrimiento Cosecha B 1987 y A 1988. Cuota de FomentoArrocero. Santafé de Bogotá.

Fedearroz. División de Investigaciones Arroceras. 1998. Arroz en Colombia,1980-97. Santafé de Bogotá.

Fedearroz. División de Investigaciones Económicas. 2000. Segundo CensoNacional Arrocero: Cubrimiento Cosecha 1999 A y B. Fondo Nacional del Arroz.Santafé de Bogotá. April.

Sanint, Luis R. 2000. Biotecnología e Investigación de Arroz: Alternativas paraAmérica Latina. Paper presented at the V Seminar "Lavoura em Evolucao", DomPedrito, Rio Grande do Sul, Brazil. June.

201

Table 3.32. Experimental production data vs. 1999 census, irrigated rice,Colombia, Semester A.

PRODUCTIVITY (Green paddy rice)Zone FEDEARROZ 50 1999-A Census

kg/ha *Cauca River Watershed 6318 5164Center 7167 7170North Coast 6309 5584Eastern Plains 4373 5360

Source: FEDEARROZ, II National Census of Rice Growers – 1999A.* Conversion factor from Green paddy field conditions to dry paddy: 0.86

Table 3.33. Productivity by zone and production system, 1999-A census,Colombia.

Zone SystemProductivity of Green Paddy **(kg/ha)

Cauca River Watershed Irrigated riceRainfed lowlands

51643562*

Center Irrigated rice 7169North Coast Irrigated rice

Rainfed lowlands55844513*

Eastern Plains Irrigated riceRainfed lowlands

54975360

Source: FEDEARROZ, II National Census of Rice Growers – 1999 A.* Coefficient of variability >25%** Conversion factor from Green paddy field conditions to dry paddy: 0.86

Table 3.34A. Productivity, Central Zone, irrigated rice, owners by sizeof rice plot, II census, 1999-A, Colombia. Green paddy rice, kgs/ha.

Small Intermediate LargeBest 10% 9573 9249 9421

Worst 10% 4794 4741 5380

Source: FEDEARROZ, II National Census of Rice Growers – 1999 A.

202

Table 3.34B. Productivity, Central Zone, irrigated rice, renters by size of riceplot, II National Census of Rice Growers, 1999-A. Green paddy rice kg/ha

Small Intermediate LargeBest 10% 8800 9331 9569

Worst 10% 4232 4937 5739

Source: FEDEARROZ, II National Census of Rice Growers – 1999 A.

Table 3.35. Distribution of the position of the irrigated rice, renters byproductivity decile, Central Zone, 1991-1998.

Worst 10% 80% Best 10%No. Semesters4 - -5 - -- 3 -- 11 -- 2 -- 10 -- 3 -- 4 -- 9 -- 12 -1 9 -1 8 31 5 13 2 -- 11 1- 5 1- 8 72 9 13 11 -- 11 22 5 -1 10 41 8 -1 12 -1 13 -- 12 2

Source: FEDEARROZ, National Rice Growers Survey, 1991-1998.

203

Table 3.36. Distribution of the position of the irrigated rice, owners byproductivity decile, Central Zone, 1991-1998.

Worst 10% 80% Best 10%No. Semesters- 2 -- 2 -- 5 -- 2 -- 10 -- 12 -- 3 -- 11 21 2 -1 4 -2 3 -- 7 11 5 1- 8 11 13 -2 11 2- 11 3- 8 31 10 -3 9 -- 8 4- 11 2- 8 33 7 1- 13 11 5 4- 12 1- 7 72 1 11 11 11 12 11 11 -- 8 53 11 -1 10 -1 11 3- 4 23 6 -1 5 -

Source: FEDEARROZ, National Rice Growers Survey, 1991-1998.

204

Publications

• Edit and publish Foro Arrocero Latinoamericano

Two annual issues have been published since the creation of FLAR in 1995. Thisbulletin constitutes a diffusion mechanism to inform the general public aboutFLAR's activities, priorities and points of view on topics related to research,training, commercialization, policies, etc related to the rice sector.

- Write editorials and articles

- Seek contributions from colleagues. Edit and produce two issues per year

• Co-Edit Arroz en las Américas

One issue was produced in 2000

• Submit articles to refereed journal

Milestone: articles submitted. Not Done. An article presented at the InternationalRice Commision of FAO (see 4.4.5 above) will be submitted to a Journal.

• Write annual reports for FLAR and for CIAT and Committee Acts

Acts for the FLAR committee meetings were written and signed by witnesses.CIAT financial office gets a copy as proof of funding commitments by partners.

• Present papers in Workshops, Conferences, Seminars, etc.

Colombia: Comalfi, XXX Congress. Invited paper on the technolgy adoption of ricein LAC. Ibague. May 10, 2000.Brazil: V Seminar on Rice Research. Invited paper on the role of biotechnology onrice research. Dom Pedrito, June 24.Colombia: Rice breeders course. The economics of rice in LAC. September 25.CIAT.

205

ANNEX 1. PRINCIPAL AND SUPPORT STAFF

Principal Staff

Lee Calvert, VirologistMarc Chatel, Plant Breeder, CIRAD-CAFernando Correa, Plant Pathologist and Project LeaderZaida Lentini, Plant BreederCésar P. Martínez, Plant BreederRafael Meneses, Visiting Scientist, I.I.A., CubaLuis R. Sanint, Agricultural Economist, Economist and FLAR Executive DirectorMichel Valés, Plant Pathologist, CIRAD-CA

Support StaffAssociates and AssistantsLuis E. Berrío, FLARJaime Borrero, GeneticsMaribel Cruz, FLARDiana Delgado, FLARMyriam Cristina Duque, BiometryJoanna Paola Dossman, Pathology, CIRAD-CAFabio Escobar, BiotechnologyJulio Eduardo Holguín, FLAR/FedearrozIván Lozano, VirologyJaime Lozano, GeneticsMaría Nelly Medina, LeaderAdriana Mora, Genetics/Anther CultureYolima Ospina, GeneticsGustavo Prado, PathologyJames Silva, BiometryMónica Triana, EntomologyEdgar Tulande, Pathology (VVC)

Visiting Scientist

Edgar Torres, DANAC, VenezuelaLuis Antonio Reyes, Fedearroz, Colombia

Secretaries

Liz Deira Arango, FLAR*María Victoria Ballesteros, FLARElizabeth Hurtado, LeaderLiliana Escobar, LeaderCarmenza Llano, Leader/Pathology*

206

Technicians and Field Support

Felix Acosta, GeneticsMiguel Acosta, FLARTomás Agrono, Genetics/Anther CultureMaría Girlena Aricapa, PathologyJesús E. Avila, Physiology (VVC)Jairo Barona, LeaderSilvio James Carabalí, GeneticsMarco Tulio Castillo, FLAREfrén A. Córdoba, EntomologyGerardo A. Delgado, GeneticsMaría Ximena Escobar, FLARJaime Gallego, GeneticsJairo García, BiotechnologyAldemar Gutiérrez, FLARJorge Ignacio Hernández, FLAR/FEDEARROZVictoria Eugenia Kury, FLARLuis Armando Loaiza, FLARVíctor Hugo Lozano, Genetics (VVC)Henry Manyoma, FLARMaría C. Martínez, VirologyFabián Mina, FLARJaime Morales, FLAR (VVC)Mauricio Morales, EntomologyRodrigo Morán, EntomologyJosé Arturo Mosquera, FLARCarlos Ordoñez, GeneticsFrancisco Ortega, Physiology/GeneticsFrancisco Rodríguez, Genetics (VVC)Humberto Rodríguez, FLARLuis H. Rosero, PathologySory H. Sánchez, GeneticsPedro Nel Vélez, GeneticsJairo Vega, FLAR (VVC)Daniel Zambrano, Pathology

* Left during 2000

Project IP-4 - CGSpace Report... · 2.A. Characterization and Genetics of Resistance to Rice Blast, Sheath Blight and Grain Discoloration 51 2.B. Characterizing and Using Partial

Documents