WORLD WHEAT FACTS AND TRENDS Global Wheat Research in a ...

WORLD WHEAT FACTS AND TRENDS Global Wheat Research in a ...

W O R L D W H E A T F A C T S A N D T R E N D SGlobal Wheat Research in aChanging World: Challengesand AchievementsP.L. Pingali, Editor

World Wheat Facts and TrendsGlobal Wheat Research in aChanging World: Challengesand AchievementsP.L. Pingali, Editor

CIMMYT ( or is an internationally funded, nonprofit scientificresearch and training organization. Headquartered in Mexico, the Center works with agriculturalresearch institutions worldwide to improve the productivity, profitability, and sustainability of maizeand wheat systems for poor farmers in developing countries. It is one of 16 similar centers supportedby the Consultative Group on International Agricultural Research (CGIAR). The CGIAR comprisesabout 60 partner countries, international and regional organizations, and private foundations. It is cosponsoredby the Food and Agriculture Organization (FAO) of the United Nations, the InternationalBank for Reconstruction and Development (World Bank), the United Nations Development Programme(UNDP), and the United Nations Environment Programme (UNEP). Financial support for CIMMYT’sresearch agenda also comes from many other sources, including foundations, development banks,and public and private agencies.CIMMYT supports Future Harvest, a public awareness campaign that buildsunderstanding about the importance of agricultural issues and internationalagricultural research. Future Harvest links respected research institutions,influential public figures, and leading agricultural scientists to underscore the wider social benefits ofimproved agriculture—peace, prosperity, environmental renewal, health, and the alleviation of humansuffering (© International Maize and Wheat Improvement Center (CIMMYT) 1999. Responsibility for thispublication rests solely with CIMMYT. The designations employed in the presentation of material inthis publication do not imply the expressions of any opinion whatsoever on the part of CIMMYT orcontributory organizations concerning the legal status of any country, territory, city, or area, or of itsauthorities, or concerning the delimitation of its frontiers or boundaries.Printed in Mexico.Correct citation: Pingali, P.L. (ed.). 1999. CIMMYT 1998-99 World Wheat Facts and Trends. GlobalWheat Research in a Changing World: Challenges and Achievements. Mexico, D.F.: CIMMYT.Abstract: This report has four parts. The first part focuses on the changing environment in which theinternational wheat research system functions in developing countries. The authors describe recenttrends in developing country wheat production against projected demand for wheat. Next, theypresent strategies for increasing productivity in favored and less favored wheat productionenvironments. Part 1 concludes with a discussion of how emerging trends and policies, such asintellectual property protection and market liberalization, are likely to affect the global wheat researchsystem and exchange of germplasm. Part 2 of the report presents new information on the historicalimpact of the international wheat improvement system, including information on the role of CIMMYTgermplasm. Part 3 discusses wheat supply and demand projections, with a special focus on Asia(including Central Asia). Part 4 presents statistics on world wheat production and consumption in anewly revised format.ISSN: 0257-876XAGROVOC descriptors: Developing countries; Wheats; Food production; Food consumption; Productionfactors; Productivity; Production increase; Economic environment; Researchinstitutions; Research projects; Diffusion of research; Innovation adoption;Economic analysis; Economic trends; Statistical dataAdditional keywords: CIMMYT; NARSAGRIS category codes:E14 Development Economics and PoliciesE16 Production EconomicsDewey decimal classification: 338.1601724

Contentsiv Acknowledgmentsv Foreword1 Part 1: Global Wheat Research in a Changing WorldP.L. Pingali and S. Rajaram1 Why Worry about Wheat Productivity?3 The Changing Nature of Wheat Productivity: The Early GreenRevolution Years6 Wheat Productivity, 1985 to the Present: New Concerns6 Enhancing Wheat Productivity in Favorable Environments7 Shifting the Yield Frontier8 Improving Yield Stability11 Enhancing Input Use Efficiency13 Enhancing Wheat Productivity in Marginal Environments14 Improving Drought Tolerance15 Improving Heat Tolerance15 Coping with Acid Soils16 Meeting the Unresolved Challenges for Marginal Environments17 Implications for Global Wheat Research17 Global Trends17 Research Strategies for Favored and Marginal Environments18 Emerging Trends in Research and Development19 Part 2: Assessing the Benefits of International Wheat BreedingResearch: An Overview of the Global Wheat Impacts StudyP.W. Heisey, M. Lantican, and H.J. Dubin27 Part 3: The World Wheat Economy, Post–2000: Focus on AsiaP.L. Pingali27 Competetiveness of Wheat Production in Developing Countries28 Special Report: Wheat for Asia’s Increasingly WesternizedDiets (P.L. Pingali and M. Rosegrant)30 Special Report: Wheat in Kazakhstan—ChangingCompetitiveness and Sources of Productivity Growth(J. Longmire and A. Moldashev)32 Targeting Wheat Research and Development Investments33 Part 4: Selected Wheat StatisticsP. Aquino, F. Carrión, and R. Calvo34 Production Statistics38 Consumption Statistics41 CIMMYT Wheat Experimental and National Average Yield,1995-9742 Wheat Area by Type of Wheat43 Prices and Input Use for Wheat44 References46 Appendix A. Regions of the World19 Wheat Research Efforts by NARSs20 Releases of Wheat Varieties22 Adoption of Improved Wheat Varieties22 Spring Bread Wheat23 Spring Durum Wheat23 Winter/Facultative Bread Wheat23 Landraces23 CIMMYT/NARS Collaboration24 Trends and Observations25 Outstanding Issues25 Varietal Replacement26 Genetic Gains in Wheat Yields and Yields in Farmers’ Fields26 The Future of International Collaboration in Wheat ImprovementResearch

ivAcknowledgmentsAcknowledgmentsThe content of this report was developed in collaboration between CIMMYT’s Economics and WheatPrograms and has benefited from the efforts of contributors from within and outside CIMMYT.Part 1 of the report was reviewed by Maarten van Ginkel and Paul Heisey. An initial draft of Part 1 waspresented at the Illinois World Food and Sustainable Agriculture Program Conference, “Meeting theDemand for Food in the 21 st Century: Challenges and Opportunities,” 28 May, 1997, University of Illinois,Urbana-Champaign, and the authors are grateful to the conference participants for helpful comments.Many CIMMYT colleagues provided information and assistance with this report, especially in questionnairedesign, data collection, and some of the data analysis. We thank Osman Abdalla, Hans-Joachim Braun,Etienne Duveillier, Paul Fox, Roberta Gerpacio, He Zhonghu, Peter Hobbs, Man Mohan Kohli, MinaLantican, Craig Meisner, Mulugetta Mekuria, Erika Meng, Mohamed Mergoum, Michael Morris, WilfredMwangi, Miloudi Nachit, Guillermo Ortiz-Ferrara, Tom Payne, Wolfgang Pfeiffer, Bent Skovmand, MelindaSmale, Douglas Tanner, Maarten van Ginkel, George Varughese, Rey Villareal, Pat Wall, and Jeff White.Piedad Moya of the International Rice Research Institure (IRRI) also assisted with the design of the WheatImpacts questionnaire.Researchers around the world worked to provide detailed information about wheat research andproduction in their home countries; their assistance has made this report a truly useful research resource.We extend our sincere thanks to Ibrahim Umar Abubakar, Amri Ahmed, Azim Akbari, Muhammad RamzanAkhtar, Julio Cesar Albrecht, Fernando Aldana, Metin Arican, Mustafa Erkan Bayram, Bedada Girma, EfremBechere, Abderrezak Belaid, M.R. Bhatta, Egon Andres Bogado, Gaetano Boggini, Antonio Maria BolañosAlomia, Basilio Borghi, S.J. Brown, Robert H. Busch, Miguel Alfonso Camacho Casas, Ivo Marcos Carraro,Brett F. Carver, Ramesh Chand, Michel Chapon, Oswaldo Chicaiza, Carlos De León, R.M. DePauw,Mahmoud Deghaies, Cantidio N.A. de Sousa, J. Gonzalo Díaz De León T., M. EL-Ghouri, Soliman El Sebai,Abdalla B. Elahmadi, Kamel Feliachi, Joao C. Felicio, Dario Fossati, Francisco de Assis Franco, MirihanGamarra Flores, J. Ganbold, Mohamed Salah Gharbi, D. Gogas, Rene W. Gomez Quintanilla, EdgarGuzman Arnez, Habib Halila, Ephrame K. Havazvidi, Hu Ruifa, IAPAR, J.S. Ijardar, B.S. Jadon, J.M. Jager,L. Jestin, M.P. Jha, A.K. Joshi, Tapio Juuti, Mesut Kanbertay, Lakshmi Kant, Mesut Keser, Fahed SalehKhatib, Frank Kiriswa, Nafees Kisana, Christof I. Kling, Gilberto Kraan, Benida Laubscher, Justin Yifu Lin,Gerard Lohan, Luis Macagno, Benvindo M. Macas, Hassan Machlab, Elizabeth J. Maeda, Abdul Majid,E. Majidi, José Marinetto Quiles, Demissie Mitiku, Raafat Abdel Hameed Mitkees, José Dilmer Moreno M.,Ashok Mudwari, M.Y. Mujahid, Mary V. Mukwavi, S. Nagarajan, G.S. Nanda, Ali Reza Navabi, Nelson Neto,Manuel Obrero Serrano, Hassan Ozcan, Ivan Panayotov, Gino Pasquali, D.R. Pokharel, Agromar O.Polidoro, Thomas Quinlivan, I. Ramírez A., V. Suryaprakasa Rao, Regassa Ensermu, Carlos Roberto Riede,Miguel Rivadeneira, S.M.A. Rizvi, Abelardo Rodriguez, José Ruedell, Juan Antonio Martín Sánchez,E. Sadeghi, G.S. Sethi, Jose J. Schvartzman, S.C. Sharma, Ali Shehadeh, M.S. Shelembi, Jag Shoran,Oscar I. Silvero Sanz, T.B. Singh, R.P. Singh, Paulo Gervini Sousa, Luis Hermes Svoboda, M. Taeb,B.R. Thapa, R.Y. Thete, Roque G.A. Tomasini, Gilberto O. Tomm, Ivan Todordv, Sencer Tumer, Hugo vanNiekerk, Jorge Velasco Lora, Ruben Verges, Jose Maria Vilela de Andrade, Reinhard von Broock, NaratWassimi, Yang Yanhua, Khaled Yassir, Mohammed Yunus, R. Zentner, and Zhao Xianlin.This report was edited by Kelly Cassaday and David Poland and designed by Eliot Sánchez. Miguel Melladodirected overall production and printing. We are grateful for their assistance.

World Wheat Facts and Trends 1998/99vForewordPrevious reports in our Facts and Trends series have examined the accomplishments of theglobal wheat improvement research system and also have forecast needs for wheattechnology in specific environments throughout the world. In this report, for the first time,we examine the global wheat improvement effort from both of these complementaryperspectives. As the title of our report indicates, we offer a comprehensive overview ofpast achievements and future strategies for wheat breeding in developing countries.The new millenium is ushering in challenges as well as opportunities for researchprograms. The Green Revolution and the emerging international wheat improvementsystem matured in conditions that were extremely conducive to agricultural research anddevelopment, but these conditions have evolved into a new set of circumstances: theglobalization of the world economy, including the world wheat economy; the trendtowards increasing protection of intellectual property; the weakening of many publicsectorresearch programs in developing countries; reduced recognition of agriculture as astimulus for economic development; and the tendency to restrict the flow of experimentalgermplasm throughout the world.As these changes in the wheat research environment become more widespread, thedeveloping world’s demand for wheat will not remain static. Demand is continuing togrow. According to recent projections by the International Food Policy Research Institute,wheat imports in developing countries will more than double between 1993 and 2020. Amajor challenge for the research and development community will be to devise strategiesfor meeting this rising demand in a more volatile research setting. Part 1 of our reportreviews projected increases in wheat demand in developing countries and describes theoptions for raising wheat productivity in favorable and less favorable wheat productionenvironments. What is proposed in Part 1 is nothing less than an international wheatimprovement agenda for the next five to ten years. Success will depend as much onadvances in science as on a less predictable factor: the support that will enable researchersto achieve their goals.As Part 2 of this report demonstrates, over the past three decades the internationalresearch system has proven to be uniquely suited to meeting the needs of poor farmersand consumers in developing countries. Data from our most recent study of the impacts ofglobal wheat research confirm that the longstanding collaboration between CIMMYT andnational wheat breeding programs in developing countries has yielded impacts on a scalethat many research efforts rarely achieve. For example:♦ The percentage of spring bread wheat releases that were CIMMYT crosses or had atleast one CIMMYT parent was even higher in 1991-97 (84% of spring wheats releasedin the developing world) than in earlier periods.♦ Sixty-two percent of the developing world’s wheat area is sown to wheats withCIMMYT ancestry.♦ Eighty to ninety percent of the developing world’s spring bread wheat area outside ofChina is planted to CIMMYT-related wheats. In China alone—the world’s largest wheatproducer—36% of the spring bread wheat area is sown to CIMMYT-related wheats.

viForewordArguably this global impact is unmatched in its extent by any other development initiative.Simply put, wheat is the most widely consumed food grain in the world, and CIMMYTgermplasm underpins the crop in the South and also contributes significantly to wheatproduction in the industrialized world (for example, in Europe, the USA, and Australia).CIMMYT has made a true contribution to global food security. Based on theseachievements, the global wheat improvement system merits major support for its researchagenda in the years to come.A main purpose of CIMMYT’s Facts and Trends series is to explore the difficult issuesrelated to maize and wheat research in developing countries. This report explores manysuch issues—including serious questions about safeguarding the flexibility of the globalwheat improvement system—but here I would like to raise two more. Although theyremain largely unspoken in the pages that follow, these questions lie at the heart of thisreport, and I would ask readers to consider them as they proceed. First, if the global wheatresearch system had not existed, what kind of world would we have inherited? Second, ifthat system should cease to operate efficiently, what legacy will we leave futuregenerations? Of all of the issues proposed for consideration in this report, these two maybe among the most relevant for directing future research and development strategies.Timothy G. ReevesDirector General

Part 1Global Wheat Research in aChanging World: Options for SustainingGrowth in Wheat ProductivityPrabhu L. Pingali and Sanjaya RajaramWorld Wheat Facts and Trends 1998/991By 2020, demand for wheat is expectedto be 40% greater than its current levelof 552 million tons (Rosegrant et al.1997), but the resources available forwheat production are likely to besignificantly lower. Viewed in this light,the challenge for increasing wheatsupplies in the developing world is asgreat today as it was three decades agoat the start of the Green Revolution.The strategies that developing countriesadopt to meet future demand for wheatwill depend a great deal on how theyare affected by the changes that aresweeping the world economy andtransforming the way we conductresearch. As global food marketsbecome increasingly integrated, thepremium once placed on food selfsufficiencyis being superseded by anemphasis on economic competitivenessand comparative advantage. Agriculturalresources are increasingly diverted fromcereal crop production to otheragricultural and nonagriculturalactivities. Research systems, national aswell as international, face decliningbudgets and uncertain futures. The freeinternational movement of germplasmand information, which was animportant force behind the GreenRevolution, is increasingly circumscribedby plant quarantine restrictions andintellectual property protection.What strategy should the global wheatresearch system pursue in this changingworld? The answer explored here is forthe research system to focus onsustaining the competitiveness of wheatproduction in developing countries. Thisgoal can be achieved through a shift inthe yield frontier, a constant drive tostabilize yields, and enhanced input useefficiency and input responsiveness inwheat varieties. The emphasis onimproving the profitability of wheatproduction should not be restricted toirrigated, favorable environments; similaropportunities ought to be explored formarginal, rainfed environments.In exploring these issues, we focus onthe roles that global wheat research andgermplasm exchange have played—andshould play—in sustaining growth inwheat productivity over the next twodecades. After discussing past trends inwheat productivity, we review potentialtechnological advances for favorable andmarginal wheat environments. Weconclude by discussing how theintegration of world food markets,economic liberalization, and greaterintellectual property protection are likelyto affect the chief source of gains inwheat productivity: the global system ofgermplasm and information exchange.Why Worry about WheatProductivity?The development and release of modernwheat varieties in the early 1960striggered the Green Revolution. The firstand most important factor contributingto the success of the wheat revolutionwas wheat itself: semidwarf, highyielding,rust-resistant wheat seed. Thesecond was the establishment of a free,unrestricted global wheat researchsystem based on the exchange ofgermplasm. The third was large-scaleinvestment in fertilizers, irrigation, andtransportation infrastructure. Lastly, thestrong political will in developing nationsto achieve food self-sufficiency,combined with a conducive agriculturalpolicy environment, also contributed tosuccess.The largest gains in productivity weremade in land-scarce countries, wherethe new seed and fertilizer technologiesfostered rapid growth in landproductivity. By the late 1970s, 40% ofthe wheat area in developing countrieswas sown to modern high-yieldingvarieties; the figure for Asia was close to70% (Table 1). By 1994, 78% ofdeveloping country wheat area wasunder modern varieties. Thecorresponding figure for Asia was 91%and, for Latin America, 92%.Between 1961 and 1994, wheat yieldsincreased at an average annual rate ofmore than 2% in all developingcountries except China and India, thetwo largest wheat producers (Pingali andHeisey 1996). Wheat yields in China andIndia grew extremely rapidly over muchof this period. Yields in India rose sharplyin the early years of the Green

2Global Wheat Research in a Changing WorldRevolution, from the mid-1960s until thelate 1970s. Although rates of growth inwheat yields in these countries havedeclined since the late 1980s, they havestill been 2% per year or more in thelatest periods for which data areavailable. Wheat yields in South and EastAsian countries other than China andIndia grew at an average rate of 2.75%annually over 1961-94, displaying muchthe same pattern as seen in India butslowing more markedly in recent years.The West Asian/North African countriesand the wheat-producing countries ofsub-Saharan Africa also experiencedrapid wheat yield growth from 1961 to1994, approximately 2.4% per annum.Latin America lagged behind with a yieldgrowth rate of around 1.8% per annum,although it started from a higher base. InWest Asia/North Africa, Latin America,and sub-Saharan Africa, rates of yieldincrease have tended to vary over time,but, except for the last region, they havebeen lower in later periods than duringthe first two decades of the GreenRevolution. (The potential implications ofslower rates of yield growth arediscussed in the next sections ofthis report.)Table 1. Percentage area planted to modern varieties of wheat a indeveloping countries, 1970–971970 1977 1983 1990 1994 1997 dSub-Saharan Africa 5 22 32 52 59 56West Asia/North Africa 5 18 31 42 57 c 61South and Southeast Asia 42 69 79 88 91 92China na na na 70 70 90Latin America 11 24 68 82 92 88All developing countries 20 b 41 b 59 b 70 78 82Source: Byerlee and Moya (1993); CIMMYT (1996); CIMMYT wheat impactsdatabase.a Excludes tall varieties released since 1965. If these varieties are included, thearea under modern varieties increases.b Excludes China.c Important countries such as Morocco and Iran not included.d Excludes unknown cultivars (i.e., those for which pedigree and origin are notknown).From the mid-1970s onwards, wheatproduction has grown at a faster pacethan population. A similar trend hasoccurred for rice. The increased yields ofthese two major cereal crops belied thewidespread fear that the world wouldrun out of food. Instead, the world hasgained access to more and cheaperfood. Since the mid-1970s, global wheatprices have declined in real terms,resulting in significant consumptionbenefits for both the urban and ruralpoor (Figure 1).Price (US$)50040030020010001970 74 78 82 86 90 94 98Marketing yearFigure 1. Wheat price (1990 constant US$).Source: Prices, ERS-USDA; deflator, InternationalMonetary Fund.The international investment in breedingmore productive wheats for developingcountries paid off handsomely. The ratesof return to wheat research havegenerally been calculated at 50% ormore for individual developing countriessince the early 1970s (Table 2).Given all of these achievements, whyshould anyone be concerned aboutwheat productivity? The wheatrevolution was good for the majority ofpoor people, but the world has changedin the three decades since farmers sowedthe first seeds of the Green Revolution.The forces that drive productivity gains inwheat are changing. They can beexpected to transform how and wherewheat is produced, and for whatpurposes; they will also change theobjectives of wheat research, the way itis conducted, and the way its productsare made available to farmers. They maywell determine whether poor farmersand consumers have enough food to eatin 2020.Table 2. Studies of returns to investment in wheat researchRate ofStudy Year Country Period return (%)Ardito–Barletta 1971 Mexico 1943–63 90Eddleman 1977 USA 1978–85 46Kislev and Hoffman 1978 Israel 1954–73 125–150Pray 1980 Bangladesh 1961–77 30–35Sundquist et al. 1981 USA 1977 97Otto and Havlicek 1981 USA 1967–79 81Yrarrazaval et al. 1982 Chile 1949–77 21–28Zentner 1982 Canada 1946–79 30–39Nagy 1983 Pakistan 1967–81 58Ambrosi and Cruz 1984 Brazil 1974–90 59–74Furtan and Ulrich 1985 Canada 1950–83 29Norton et al. 1987 Peru 1981–2000 18–36Evenson and da Cruz 1989 Brazil 1979–88 110Byerlee 1990 Pakistan 1978–87 16–27Evenson and McKinsey 1991 India 1972–84 51Morris et al. 1992 Nepal 1960–90 75–84Macagno and Gómez Chao 1993 Argentina 1967–92 42Source: Echeverría, cited by Morris et al. (1994); CIMMYT (1993).

World Wheat Facts and Trends 1998/993The next two sections of this reportdescribe some of the forces influencinggains in wheat productivity and how theyhave changed over the years. Tounderstand these changes and theirimplications for future gains in wheatproductivity, we review the progress ofthe Green Revolution in favorable(irrigated) and marginal (dryland)environments, and then we examinehow growth in yield, trends in input use,and types of production incentives havechanged since the wheat revolutionbegan. For an overview of the variouswheat production environments, see thebox, “Wheat Mega-EnvironmentsDefined,” p. 4.The Changing Nature of WheatProductivity: The Early GreenRevolution YearsThe first gains in wheat productivity wereseen in irrigated and high-rainfallenvironments, the so-called favorableproduction environments. By 1977, 83%of the wheat area in these environmentswas planted to modern, high-yieldingwheats. By 1990, this figure was close to100% (Byerlee and Moya 1993).The irrigated wheat-growingenvironments exhibited substantial yieldgrowth. In northwestern Mexico, forinstance, average farm yields had risenfrom 2.2 t/ha in 1960 to 6.0 t/ha in1998, almost a three-fold increase(CIMMYT Economics Program data).Similarly, in the Indian Punjab, averageyields rose from 1.5 t/ha in 1960 to 3.5t/ha in 1989 and 4.2 t/ha in 1999 (Sidhuand Byerlee 1992; Punjab AgriculturalUniversity, pers. comm.).During the first three decades aftermodern varieties were introduced, theyield potential of subsequent generationsof modern varieties in favorableenvironments rose by 1% per annum(Byerlee and Traxler 1995) (Table 3). Theinitial yield boost of 35-40% on farmers’fields was followed by a period of lessdramatic but nonetheless steady yieldgrowth, during which the second- andthird-generation varieties with higheryield potential and increased diseaseresistance replaced the original improvedvarieties (Byerlee and Morris 1993).The rainfed environments, marginal forwheat production, also benefited fromtechnological change. First theybenefited from a spillover of newvarieties from the irrigated environments;later, from new varieties adapted torainfed conditions, especially varietieswith improved drought tolerance. By1990 approximately 44% of all wheatvarieties released were bred specificallyfor dryland environments (Table 4).Table 3. Experimental evidence on rates of genetic gain in yields in spring wheat resultingfrom the release of new varieties (yield maintenance effect not included)Rate ofEnvironment/location Period gain (%/yr) Data sourceIrrigatedSonora, Mexico 1962–75 a 1.1 Fischer and Wall (1976)1962–83 a 1.1 Waddington et al. (1986)1962–81 a 0.9 P. Wall, CIMMYT b1962–85 a 0.6 Ortiz–Monasterio et al. (1990)1962–89 a 0.7 K.Sayre, CIMMYT bNepal 1978–88 a 1.3 Morris, Dubin, and Pokhrel (1994)Northwest India 1966–90 a 1.0 Jain and Byerlee (1994)Pakistan 1965–82 a 0.8 Byerlee (1993)Sudan 1967–87 0.9 Byerlee and Moya (1993)Zimbabwe 1967–85 a 1.0 Mashiringwani (1989)RainfedParaguay 1972–90 1.3 M.Kohli, CIMMYT bVictoria, Australia 1850–1940 0.3 O’Brien (1982)1940–81 0.8New South Wales, Australia 1956–84 0.9 Antony and Brennan (1988)Western Australia (low rainfall) 1884–1982 0.4 Perry and D’Antuono (1989)Central India 1965–90 0.0 Jain and Byerlee (1994)Acid soils (rainfed)Rio Grande do Sul, Brazil 1976–89 3.2 Byerlee and Moya (1993)Paraná, Brazil 1969–89 2.2 Byerlee and Moya (1993)Source: Byerlee and Traxler (1995).Note: Regression results and data available from authors.a Semidwarf varieties only.b Unpublished data.Table 4. Distribution of wheat varieties released in developing countries, by wheat type andecological niche, 1966–97Percent recommended for:Well-watered/irrigated Dryland BothWheat type/region 1966–90 1991–97 1966–90 1991–97 1966–90 1991–97Spring bread wheat 37 58 45 18 19 24Sub-Saharan Africa 36 72 39 8 25 20West Asia/North Africa 57 51 14 42 29 7South and East Asia 70 58 14 9 16 33Latin America 13 56 71 15 15 29Winter bread wheat 26 46 56 32 18 22Spring durum wheat 43 41 34 30 20 29All 37 53 44 23 19 24Source: Byerlee and Moya (1993); CIMMYT wheat impacts database.

4Global Wheat Research in a Changing WorldWheat Mega-Environments DefinedIn the mid-1980s, CIMMYT refinedits definitions of the environmentsto which it targets its wheatgermplasm by grouping them into“mega-environments” (MEs),which were developed based on amixture of plant, disease, soil,climatic, and socioeconomiccharacteristics. An ME is a broad,frequently transcontinental but notnecessarily contiguous area occurringin more than one country. It isdefined by similar biotic and abioticstresses, cropping systemsrequirements, consumer preferences,and, for convenience, volume ofproduction. The MEs are useful fordefining breeding objectives becauseeach one encompasses millions ofhectares sharing a certain degree ofhomogeneity for wheat production(Dubin and Rajaram 1996).Characteristics of wheat mega-environments (ME)YearbreedingWheat type Latitude Moisture Temperature Representative beganand ME (degrees) regime a regime b Sown Breeding objectives c,d locations or regime at CIMMYTSpring wheatME1 d 40 High rainfall, Severe cold Autumn Resistance to LR, SR, PM, winterkill, Lovrin, Romania 1986short seasonsproutingME12 >40 Low rainfall Severe cold Autumn Resistance to winterkill, drought, Ankara, Turkey 1986YR, buntsSource: Rajaram and Van Ginkel (1996).a Rainfall refers to just before and during the crop cycle. High = > 500 mm, low = < 500 mm.b Refers to the mean temperature of the coolest month. Hot = >17 C; temperate = 5-17 C; moderate cold = 0-5 C; severe cold = -10-0 C.c Factors additional to yield and industrial quality. SR = stem rust, LR = leaf rust, YR = yellow (stripe) rust, PM = powdery mildew; BYD = barley yellow dwarf.d Further subdivided into: (1) optimum growing conditions, (2) presence of Karnal bunt, (3) late planted, and (4) problems of salinity.

World Wheat Facts and Trends 1998/995By 1992, 12 MEs involving springwheats (ME1 to ME6), facultativewheats (ME7 to ME9), and winterwheats (ME10 to ME12) had beendefined (see table). Spring wheatMEs cover almost 95 millionhectares in the developingcountries, excluding China, whilefacultative/winter wheat MEs coveralmost 25 million hectares (Dubinand Rajaram 1996). Of the 12 MEs,by far the most important in areaand production is the irrigatedspring wheat environment, where40% of the developing world’swheat is produced. Spring wheat—durum wheat and bread wheattogether—makes up 75% of totalwheat production in developingcountries (including China)(CIMMYT wheat impacts database).For this reason, the discussion inthis report emphasizes the springwheat MEs. The characteristics ofthe six spring wheat MEs used bythe CIMMYT Wheat Program aresummarized in the table. Of thesesix MEs, ME1 and ME2 areconsidered the favorable wheatenvironments, and ME3 to ME6 arethe marginal ones.Byerlee and Traxler (1995) reported thatover the past few decades the yieldpotential of drought-tolerant varietieshas risen at a modest 0.5% per annum(Table 3).Rainfed area sown to modern wheatvarieties has risen steadily since the midtolate 1970s (see Figure 2 forexamples). Yield gains from theadoption of modern varieties in thesemarginal environments have beenmodest (Byerlee and Morris 1993), butan observable drop in yield variability hasPercent area MVs12010080604020IrrigatedPakistanRainfed01967 69 71 73 75 77 79 81 83 85 87 97Percent area MVs12010080604020IrrigatedRainfedSyria0196769 71 73 75 77 79 81 83 85 87 89 97resulted from the release of wheatgermplasm possessing improvedtolerance to drought and other abioticstresses. Using time-series data for 57countries, Singh and Byerlee (1990)found that variability in wheat yieldsdeclined by an average of 23% from1951-65 to 1976-86. Table 5 shows thatthe coefficient of yield variabilitydropped by 9% in the drylandtemperate environments and by 16% inthe tropical environments. Wheat yieldvariability dropped significantly in theirrigated environments as well.Percent area MVs12010080604020IrrigatedRainfedArgentina01971 73 75 77 79 81 83 85 87 8997Figure 2. Adoption of modern wheatvarieties in different moisture zones.Source: Byerlee and Moya (1993) and CIMMYTwheat impacts database.Table 5. Changes in the variability of wheat yields in selected groups of countries, 1951–86Coefficient of variation (CV)Number of around trend (%) Percent change in CV,countries 1951–65 1966–75 1976–86 1951–65 to 1976–86Irrigated and wellwateredtemperate 36 10.6 8.5 6.4 –38Dryland temperate 15 16.2 15.2 14.4 –9Tropical 6 22.6 23.6 18.9 –16All countries 57 13.3 11.8 10.1 –23Source: CIMMYT (1991).

6Global Wheat Research in a Changing WorldWheat Productivity, 1985 to thePresent: New ConcernsAs noted, for the first time since the startof the Green Revolution, seriousconcerns about future wheat supplieshave emerged. Newer generations ofhigh-yielding varieties continue to exhibithigher yield relative to earlier varieties,but the rate of growth in yield potentialhas slowed considerably in the mostrecent decade (Sayre 1996).In environments favorable to wheatproduction, the economically exploitablegap between potential yields and yieldsachieved on farmers’ fields has beenreduced considerably over the past threedecades. 1 In other words, in these areasthe cost of marginal increments in yield,given existing technologies and policies,could exceed the incremental gain. Thecost is high not only in terms ofincreased use of inputs such as fertilizer,fuel, and water, but also in terms ofincreased management and supervisiontime for achieving more efficient inputuse (Pingali and Heisey 1996;Byerlee 1996).Productivity growth in the irrigatedenvironments has also been affected bythe decline in irrigation investments sincethe early 1980s. Reduced expenditureson irrigation have not only limited theexpansion in irrigated area but havenegatively affected the maintenance ofinfrastructure (Rosegrant and Pingali1994). Decades of poor watermanagement have caused large tracts ofirrigated land to be abandoned orcultivated at low productivity levelsbecause of salinization. The IndianPunjab, Pakistani Punjab, Yaqui Valley innorthwestern Mexico, and the irrigatedwheat areas in the Nile Valley exhibitvisible signs of land degradation fromsalinity buildup. Salinity is thought toaffect nearly 10 million hectares ofwheat in developing countries.In marginal environments, farmers willface increasing difficulty in capturingfurther productivity gains. Factors relatedto plant variety and crop managementare the basis for this prognosis. Thegains farmers receive from using wheatvarieties originally developed for irrigatedenvironments (i.e., spillovers or spillins)will decrease. Additional genetic gainswill have to come from breeding effortstargeted specifically to the uniquecharacteristics of marginal environments.Furthermore, farmers in less favorableproduction areas cannot fully exploit theyield potential of the varieties they growuntil they have adequate managementpractices, which presumably would differfrom practices used in irrigatedenvironments. Outside of the AnatolianPlateau of Turkey and rainfed wheatareas of Argentina, little effort has beendevoted to developing wheatmanagement practices for marginalenvironments.Finally, farm-level incentives forincreasing productivity growth in eitherfavored or marginal areas are muchweaker today than during the first twodecades of the Green Revolution. Outputprice protection and input subsidies, thehallmarks of Green Revolutionagricultural policy, have been removed orare being removed in most developingcountries. Gradually, a global movementtowards free trade and economicintegration is replacing the closedeconomy,food-self-sufficiency model1 Sidhu and Byerlee (1992) provide evidence for the Indian Punjab; similar evidence is available fornorthwestern Mexico (CIMMYT Economics Program data).that made the Green Revolution sosuccessful. The new paradigm for foodpolicy is self-reliance and comparativeadvantage rather than self-sufficiency.Related changes in the global wheateconomy include the liberalization ofwheat imports and a long-term declinein real wheat prices, coupled with risingcosts of inputs, especially labor, land,and water. Investments in strategies forincreasing growth in the productivity ofcereal crops, including wheat, will beaffected significantly by these changes inparadigm and policy.Enhancing WheatProductivity in FavorableEnvironmentsGiven the trends we have just described,the competitiveness of wheat farmingwill depend on opportunities fordramatically reducing unit costs ofproduction. This can be achieved eitherby shifting the yield frontier and/orincreasing input-use efficiency. Researchshould ensure that the rate of increase inwheat yields surpasses the rate ofincrease in input use, so the unit cost ofproduction will decline. For wheatfarming to remain profitable,technological change ought to ensurethat production costs per ton of wheatare falling at least at the same rate asthe real price of a ton of wheat.The following sections describe geneticand crop management technologies thatCIMMYT and national research programsare developing to sustain the profitabilityof wheat production in the irrigated andhigh-rainfall areas of developingcountries. Genetic improvement researchis targeted towards shifting the yieldfrontier, reducing yield variability causedby biotic and abiotic stresses, and

World Wheat Facts and Trends 1998/997increasing input-use efficiency. Cropmanagement research focuses onimproving the partial factor productivity(output per unit of input) of land,fertilizer, water, power, and human labor.Shifting the Yield FrontierOver decades, breeders have learnedthat breeding/selection methods andexperimental field designs contribute tobreeding progress but in themselves arenot powerful enough to overcomeinferior germplasm. What matters mostin plant breeding is germplasm(Rasmusson 1996). Circumstantialevidence from practical breedingexperience supports the idea thatbreakthroughs in yield potential havelargely resulted from wider utilization ofplant genetic resources (Rajaram and vanGinkel 1996).Researchers at CIMMYT are investigatingtwo major strategies, distinct butinterrelated, for achieving a dramaticshift in the wheat yield frontier: changesin plant architecture and the exploitationof heterosis (i.e., the development ofhybrid wheat). Significant progress hasbeen made in both areas, and theprospects for a substantial shift in theyield frontier within the next decadeare good.Changes in wheat plant architecture.Wheat plant architecture has evolvedover the past 45 years, primarily tosustain growth in genetic yield potential.Efforts to develop the first semidwarfwheat varieties in Mexico sought toreduce lodging by reducing plant height,thereby enabling plants to respond moreeffectively to fertilizer. The dwarfinggenes possessed by semidwarf wheatsalso positively affected yield by allowingmore tillers to survive and thusincreasing biomass. Controlledexperiments showed that wheat yieldsincreased by at least 15% in semidwarfwheats compared to yields of tallvarieties (Hoogendoorn et al. 1988).Farm-level yields increased substantiallymore, however, because of the modernvarieties’ enhanced responsiveness tofertilizer. The new varieties also increasedthe prospects for intensified land usebecause they carried genes forphotoperiod insensitivity and earlymaturity (for example, in South Asia thenew wheats matured quickly enough forfarmers to add them to their ricecropping systems). Modern semidwarfwheats were also fortified geneticallywith durable stem and leaf rustresistance to ensure high yields.In the late 1960s, the first set ofsemidwarf wheats was hybridized withwinter wheats, producing high-yieldingspring x winter wheat varieties. The mostsuccessful of these crosses developed byCIMMYT is Veery. Compared to othermodern wheats, Veery lines have asignificantly higher grain number andbiomass. Although the wheat kernels ofVeery wheats are smaller than those ofother contemporary varieties, the drop insize is less than proportional to theincrease in grain number, hence thesignificant positive effect on yields. Veerylines yield 10–15% more than the firstgeneration of modern semidwarfvarieties. The Veerys and their progenies(including the Kauz, Attila, Pastor, andBaviacora lines) respond not only togood growing conditions but alsodemonstrate superior performanceunder a number of abiotic stressconditions, such as drought and heat,and are more efficient at using nitrogenand phosphorus.The next step in changing plantarchitecture is to design wheats withthicker stems, fewer tillers, larger heads,and a higher number of grains without acommensurate decline in grain weight.This change should increase yields byimproving the harvest index and inputuse efficiency. Through 20 years of prebreeding,genetic manipulation, andcountless recombinations, CIMMYTbreeders have developed a new wheattype called Buitre, which has a robuststem, long head (>30 cm), multiplespikelets, florets with large glumes, alarge leaf area, and broad leaves.Presently these positive traits arecounterbalanced by an unknownphysiological imbalance or disorder thatresults in a low number of tillers, largelysterile heads, mostly shriveled grains, andhigh susceptibility to leaf and stripe rusts(Rajaram, Singh, and van Ginkel 1996).CIMMYT is improving the Buitre ideotypeto develop a plant that has a largenumber of spikes, a slightly reducedhead size, and completely restoredfertility. Such genetic stocks couldpotentially increase yields by 10–15%above those of the Veery descendants insubtropical environments (Rajaram,Singh, and van Ginkel 1996). In subtemperateenvironments, where thegrain filling period is longer, yields mayincrease by 20–30%. To achieve theseyields, better soil fertility managementand well-timed irrigation managementare needed. The probability of success ingenerating these varieties is high, andthey could be available for farmers totest in 10 years. This new generation ofwheat is primarily targeted for irrigatedenvironments where the economicallyachievable yield increases from presentdaycultivars grown under optimalmanagement are small.

8Global Wheat Research in a Changing WorldHybrid wheat. A hybrid is the firstgenerationoffspring of a cross betweengenetically different parents of a plantspecies. Hybrids exploit a phenomenoncalled “hybrid vigor” or “heterosis,”which is best described as the tendencyof offspring to perform better thaneither of their diverse parents. Breedershave developed hybrid maize and rice,but the advent of hybrid wheat has beenslower for several reasons. Jordaan(1996) and Picket (1993) have attributedthis lag to the high cost of malesterilization and seed production,hybrids’ limited heterotic advantage, andtheir lack of agronomic, quality, ordisease resistance advantages overinbred lines. Hybrids have thus failed toperform better than successivegenerations of modern cultivars,especially given that yields of thosecultivars have risen by 1% per annumover the past three decades (Byerlee andMoya 1993). Results, however, fromSouth Africa (Jordaan 1996) and acrosslocations in the southern Great Plains ofthe US (Peterson et al. 1997) haveshown that the mean grain yield ofhybrids was significantly higher than thatof pure lines. A 12–17% advantage inhybrid yield over the leading checkcultivars has also been reported(Edwards 1995 and Monsanto 1997,cited in Çukadar et al. 1997).Interest from the national programs ofChina and India led CIMMYT to reinitiateits hybrid wheat program in1996. The CIMMYT program aims todevelop a practical hybrid seedproduction scheme and identify hybridswith high yield potential under favorableirrigated conditions. Three otherdevelopments have made it worthwhileto reassess hybrids: improvements inchemical hybridization agents (CHA), theemergence of the new Buitre plant typewith large spikes (described previously),and advances in biotechnology(discussed later). Chemical hybridizationagents induce male sterility and allow forrapid development of hybrids. They werefirst considered unacceptably toxic, untilrecent technological advances reducedtheir toxicity. The US EnvironmentalProtection Agency has approved the useof one CHA, Genesis®, for hybrid wheatdevelopment, making hybrid researchmore attractive. Private ownership of thechemical may impede its wideruse, however.Current efforts at CIMMYT are targetedtowards exploiting heterosis anddeveloping hybrids using CHAs.Together, increased grain filling andheterosis could shift the yield frontier ofthe new plant germplasm by 15–20%.The breeding program’s ability todevelop heterotic combinations of thenew plant material will be establishedwithin 3–5 years, although the actualdevelopment and dissemination ofhybrids will take perhaps 10 years. An exante assessment of the economic, social,and institutional impediments, includingintellectual property rights, to hybridwheat development and dissemination isurgently required.Improving Yield StabilityAside from breeding for yield potential,research programs should simultaneouslyinvest in breeding for resistance to bioticstresses, quality characteristics, andtolerance to abiotic stresses in theregions they target. Safeguarding andprotecting yield potential throughbreeding for resistance to biotic andabiotic stresses is an important basis forstable yields and wide adaptation. Inbreeding wheat for high-potentialenvironments, researchers shouldemphasize developing resistance tobiotic stresses (e.g., diseases and pests).Many breeding programs fail to deliversuitable high-yielding cultivars to farmerssimply because the cultivars aresusceptible to emerging pathogens orbecome susceptible soon aftertheir release.Conventional breeding for improvedhost-plant resistance. In its research todevelop desirable levels of disease andinsect resistance, CIMMYT combines“shuttle breeding,” “hot spot”screening, and multilocational testingwithin Mexico and abroad to obtainresistance to multiple diseases for thedifferent wheat mega-environments(Dubin and Rajaram 1996). “Shuttlebreeding” is a method in whichgenerations of plants from the samecrosses undergo alternate cycles ofselection in environmentally contrastinglocations to combine desirablecharacters. “Hot spots” are locationswhere significant variability for apathogen exists. In these locations,plants can be screened in the presenceof the broadest possible range ofvirulence genes and their combinations.This screening, together withmultilocational testing, increases theprobability of developing cultivars withdurable resistance.During the last 40 years, considerablegenetic progress has been achieved indeveloping host-plant resistance inwheat to leaf rust, stripe rust, septorialeaf blotch, fusarium head scab, andbacterial leaf streak in high-rainfall areasof the tropical highlands. Some progresshas also been made in achievingresistance to diseases of secondaryimportance, such as barley yellow dwarf,

World Wheat Facts and Trends 1998/999tan spot, and septoria nodorum blotch(Dubin and Rajaram 1996). For decades,breeders have sought to ensure thateach new elite cultivar and breeding linepossesses a wide range of genes forresistance to pests and diseases.Breeding programs are challenged by thefact that genetic variation is not confinedto the host plant but is also a feature ofthe pest or parasite. With certain typesof resistance, selection for a resistantpopulation of the host is closely followedby natural selection of the pest orparasite for those variants that canovercome the resistance. To keep aheadof evolving and changing pestpopulations, breeders must constantlyaim to increase genetic diversity againstpests and develop cultivars with moredurable resistance.Wide crossing. Much of the recenteffort to enhance the diversity of wheatbreeding pools and improve yield stabilitycenters on wide crossing, a technique forintroducing desirable genes from wildspecies into cultivated varieties. Forwheat, the Triticum grass species are animportant source of enhanced geneticdiversity, especially goat grass (Triticumtauschii) (Mujeeb-Kazi and Hettel 1995).Perennial grasses such as T. tauschii canprovide a wide range of geneticresistance/tolerance to several biotic/abiotic stresses, appear to be a potentsource of new variability for importantyield components, and enhance yields inbread wheat.At CIMMYT, wide crosses between elitedurum wheats and T. tauschii yieldedsynthetic hexaploid wheats, which areused extensively in the bread wheathybridization program. CIMMYT hasproduced 620 synthetic hexaploids, andan elite subset of 95 has been preparedand partially characterized formorphological traits, yield components,growth, and resistance/tolerance toseveral biotic and abiotic stresses. Allsynthetic hexaploids are cytogeneticallystable.Seed size and weight tend to be verymuch above average in the synthetics.Major gains in yield and yield stabilityshould come from using synthetichexaploids as parent material in theproduction of hybrid wheat, primarilybecause they will help limit the effects ofbiotic and abiotic stresses. Underconditions in Mexico, synthetichexaploids have shown diversity forresistance and/or tolerance to more thanten stresses. For example, resistance toKarnal bunt, septoria tritici blotch,fusarium head scab, and spot blotchfrom synthetic hexaploids is used forbread wheat improvement based ondisease screening data obtained overseveral years. The resistant synthetichexaploids have been crossed with elitecultivars, and the resulting lines expressdiversity for resistance to these diseases.Synthetic hexaploids have alsocontributed to diversifying the geneticbase of resistance/tolerance to importantabiotic stresses such as heat, drought,waterlogging, and frost at flowering.Contributions of biotechnology tobreeding for yield stability. Newbiotechnology tools have the potentialfor considerably increasing theeffectiveness and efficiency of wheatbreeding programs, providing insightsinto the genetic control of key traits andmarkers for manipulation, methods forintroducing novel sources of geneticvariation, and methodologies forspeeding up the breeding cycle (Snape1996). In the short to medium term,however, the greater contribution ofbiotechnology will not be to raise yieldsso much as to improve yield stability bygenerating plants with improvedresistance to pests and tolerance toabiotic stresses.By facilitating gene selection, molecularmarker technology increases theefficiency with which specific desirablegenes are combined into improvedbreeding lines. Markers are especiallyuseful in breeding for resistance tononendemic diseases and pests or forincorporating genes with overlappingeffects that can contribute to complexand more durable resistance. They arealso desirable for working on traits thatare expensive to measure, for certainquality-related characteristics, and fortraits that are heavily influenced byenvironmental fluctuations. Markerassistedselection may also help identifyheterotic groupings of parentgermplasm, making the rapiddevelopment of hybrid wheat linespossible (Jordaan 1996).Sorrells (1996) has observed thatmarkers are generally cost effective foronly those traits that are difficult orcostly to evaluate and are controlled by afew genes. The costs of developing andusing molecular markers are stillrelatively high compared to the costs ofconventional selection methods, but theyare gradually declining.Although markers should play anincreasing role in resistance breeding,there are few examples of markers usedfor selection in applied wheat breedingprograms (McIntosh 1998). CIMMYTuses marker-assisted selection to develop

10Global Wheat Research in a Changing Worldresistance to barley yellow dwarf (BYD),one of the most widespread anddamaging viral diseases of wheat. Linescontaining genes for BYD resistanceintrogressed from a wild relative ofwheat (Thinopyrum intermedium) arebeing tested in the greenhouse and thefield, but resistance levels are not yetsatisfactory. Thinopyrum-derivedresistance genes may also be combinedwith known and widespread BYDtolerancegenes from other sources, inthe hope that the cumulative effects ofthese genes will confer an acceptablelevel of BYD resistance. Preliminaryresults indicate that resistance andtolerance can be combined, resulting infewer disease symptoms and reducedyield losses.Plant genetic engineering (croptransformation) is a process foridentifying and incorporating genes forvaluable traits into a specific crop. Thesegenes, which may originate from anyliving organism, virus, or even chemicalsynthesis, are introduced into recipientcells and, preferably throughincorporation into the chromosomalDNA, create a new plant with thedesirable traits. Transformation offersbreeders new opportunities to improvethe efficiency and stability ofproduction and increase the utility ofagricultural crops.Agronomically viable levels of durableresistance in crops against a relativelybroad range of fungi, such as thepathogens that cause rust diseases,might be achieved using transformationapproaches that include insertion andexpression of genes encoding inhibitorsof fungal enzymes or known antifungalproteins. Antifungal genes have beentransformed into a range of crop species,including tobacco, tomato, canola, andrice, and the insertion of such genes intowheat would provide plant breederswith additional sources of resistance.Wheat transformation research atCIMMYT has two objectives: first, toobtain durable resistance to a range offungal pathogens, and second, to raisethe level of expression of seed storageproteins in wheat endosperm throughthe introgression of genes for highmolecular-weightglutenin subunits. Theunique bread-making characteristic ofwheat flour is closely related to theelasticity and extensibility of the glutenproteins stored in wheat starchyendosperm.Before transformation technology can befully utilized to develop new cultivars,and before those cultivars can beadopted at the farm level, severalprocedural and policy issues need to beaddressed. A major uncertainty is theregulatory framework governing therelease of genetically engineeredproducts, particularly because ofgrowing concern about the public healthand environmental safety implications ofgenetically modified organisms (GMOs).Substantial work has to be done toclarify the environmental and humanhealth impacts of GMOs. Farmers’eventual access to genetically modifiedcultivars also depends on a large-scalepublic awareness program addressingthe concerns of everyone involved in thecurrent debate on GMOs. Intellectualproperty protection and plant breeders’rights may further limit access tomaterials developed throughbiotechnology. In developing countries,substantial policy reforms that allow forthe legal enforcement of intellectualproperty rights should be in place beforethe products of biotechnology researchare widely available.CIMMYT’s strategy to achievedurable rust resistance in wheat. Asnoted, the long-term aim of wheatimprovement at CIMMYT is to producegermplasm characterized by stable, highyields and possessing durable resistanceto various parasite complexes. Geneticmanipulation of resistance genes overthe last 40 years has enabled wheatplants to hold their own againstmutating pathogens of some previouslydevastating diseases, including Pucciniagraminis tritici (stem rust) and P.recondita tritici (leaf rust), resulting inpartial to good stability of resistance. Thecurrent importance of the rust diseases inseveral spring wheat megaenvironments2 is shown in Table 6.In the 1950s and early 1960s, breedersgenerally sought to incorporate thehypersensitive (race-specific) 3 type ofresistance into wheat. This approach wasattractive because the cleanliness of theTable 6. Current importance of leaf, stem,and stripe rust diseases of wheat in springwheat mega-environments (MEs)ME Leaf Stripe StemME1: Favorable High Medium LowME2: High rainfall Medium High LowME3: Acid soils High Low HighME4: Semiarid Medium Medium LowME5: Tropical High Low LowME6: High altitude High Low Low2 One or more of the rust diseases (leaf rust, stem rust, or stripe rust) are the most economicallyimportant diseases in many wheat production environments (Byerlee and Moya 1993).3 Race-specific resistance, controlled by major genes having large effects, is easily detected with a specificrace of pathogen (pathotype). In wheat rust pathosystems, this resistance is recognized by itscharacteristic low infection type. Numerous genes for race-specific resistance are now known and havebeen catalogued by McIntosh, Hart, and Gale (1995). Detection of these genes requires either seedlingevaluation or testing at post-seedling growth stages.

World Wheat Facts and Trends 1998/9911crop was preserved and it was simple toincorporate resistance into improvedgermplasm. However, the protectionafforded by race-specific genes (orcombinations of them) eroded quickly,leaving scientists to look for alternativeapproaches.In the late 1960s and 1970s, the conceptof general (race-nonspecific) 4 resistanceand its application in crop improvementwas revived (Caldwell 1968) and widelyused to breed for rust resistance inwheat (Borlaug 1968, 1972; Caldwell1968). This concept, commonly knownas slow rusting, 5 has dominatedCIMMYT’s efforts to breed for leaf rustresistance in bread wheat for more than25 years and is being used to developresistance to yellow rust as well.Since wheat varieties derived fromCIMMYT germplasm are grown overlarge areas and may be exposed topathogens under conditions that favorthe development of disease, CIMMYThas sought to use germplasm sources forrust resistance that are as diverse aspossible. The flow of germplasm to andfrom the bread wheat improvementprogram is continuous, and CIMMYTscientists are constantly in touch withcollaborating research programs toensure this exchange. Althoughmultilocational testing is not a perfectsystem for identifying diverse sources ofresistance, it helps to confirm theexistence of genetic diversity inCIMMYT’s germplasm, to understand thegenetic basis of resistance, and to judgethe performance of regionally importantadvanced breeding materials againstpathogen variation occurring in diverseenvironments. Genetic studies suggestthat wheat genotypes that are resistantto a given rust disease in manylocations—as indicated by low averagecoefficients of infection (ACI)—oftencontain multiple major or minorresistance genes.Table 7 shows the phenotypic diversityfor resistance to leaf, stripe, and stemrusts of 280 advanced bread wheat linesincluded in the 24 th International BreadWheat Screening Nursery (IBWSN).Marked differences among phenotypessuggest the existence of different groupsof varieties whose response to rust couldbe subject to different genetic controlmechanisms. Progress in breeding for leafand stem rust resistance has beensignificant, given that approximately60% of the IBWSN entries have ACITable 7. Phenotypic diversity in 280 advanced lines of the 24 th International Bread WheatScreening Nursery classified into average coefficients of infection (ACI) for three rustdiseases in international multilocational testing, 1990Number of entries in ACI classesDisease Number of locations 0–5 5.1–10 10.1–20 20.1–30 30.1–40 > 40Leaf rust 39 168 78 32 2 0 0Stripe rust 16 26 62 104 71 15 2Stem rust 15 162 52 60 4 2 04 Race-nonspecific resistance operates against all pathotypes. This type of rust resistance, based on theadditive interactions of several genes having minor to intermediate effects, is usually complex andrelatively difficult to identify.5 Slow rusting is a type of resistance in which rust develops slowly, resulting in intermediate to lowdisease levels against all pathotypes (Caldwell 1968). Partial or incomplete resistance is characterized bya reduced rate of epidemic development despite a high or susceptible infection type (Parlevliet 1975).values of less than 5. However, muchmore progress remains to be achieved instripe rust resistance, since only about10% of the entries had such lowACI values.Genetic diversity and durability are thetwo most important features of thedisease resistance that CIMMYTresearchers seek to incorporate into theirwheat lines. Because proof of enduringdisease resistance comes only aftersupposedly resistant cultivars aredeployed over a large area, it is importantthat these cultivars be genetically diverseto ensure against vulnerability. Data onthe historical performance of cultivarswith different kinds of resistance canhelp identify sources of durableresistance. Genetic analyses of thisresistance could aid in its directedtransfer and in the search for genes thatcould contribute to new genecombinations for durable resistance.Enhancing Input Use EfficiencyThe profitability of wheat productionsystems in favorable environments doesnot depend on wheat improvementresearch alone. Crop managementtechnologies that enhance input useefficiency are essential. Achievingsustained growth in wheat productivitywhile at the same time conserving theresource base will require wheatproduction increases to be achieved witha less than proportionate increase ininput use. Changes in fertilizerapplication practices, especially thetiming and method of application, couldsignificantly reduce nutrient losses andimprove plant nutrient uptake. Efficiencygains made through improvements infertilizer management may contribute toa reduction in the overall fertilizerrequirements for sustainingproductivity growth.

12Global Wheat Research in a Changing WorldToday, nearly all strategies to implementsustainable crop management practicesinvolve efficient fertilizer applicationmethods, integrated pest managementpractices, 6 conservation or zero tillage,and crop residue management (Sayre1998). Improved land managementpractices also contribute to enhancedefficiency of water use. Mechanizationof land preparation as well as harvestand post-harvest activities will lead toimproved labor productivity and theensuing gains in labor use efficiency.Improved nitrogen use efficiency.Since the 1950s, successive CIMMYTwheat releases have shown eitherincreasing yield stability, higher meanyields, or both (Traxler et al. 1995).Compared to tall varieties, leadingvarieties based on CIMMYT germplasmhave also required smaller and smalleramounts of land and nitrogen toproduce the same amount of wheat(Figure 3; CIMMYT 1996). Theimplication is that these varieties willenable farmers to release more land foralternative uses and avoid possibleoverapplication of nitrogen.N (kg for 5 t wheat)40030020010001970s1980s1960sThe adaptation and performance ofCIMMYT bread wheats under low levelsof nitrogen have been questionedbecause these wheats were developedunder medium-high levels of nitrogenfertility. To address this doubt, CIMMYTconducted a study to: 1) compare theperformance of a set of tall and a set ofsemidwarf CIMMYT cultivars widelygrown by farmers in northwesternMexico under low and high nitrogenfertility levels; 2) measure the geneticprogress in grain yield and nitrogen useefficiency (NUE); and 3) evaluate thecontribution of nitrogen uptakeefficiency (UPE) and utilization efficiency(UTE) to NUE (Ortiz-Monasterio et al.1997). Genetic gains in both grain yieldand NUE from 1950 to 1985 were 1.1,1.0, 1.2, and 1.9% per year on a relativebasis, or 32, 43, 59, and 89 kg/ha/yr onan absolute basis when plants wereprovided with 0, 75, 150, and 300 kg/haof N, respectively. Progress in NUEresulted in improved uptake andutilization efficiency. The relativeimportance of these two componentswas affected by the level of appliednitrogen, however. A shift in the wheatyield frontier could lead to further gainsin uptake and utilization efficiency.0 0.5 1 1.5 2Land area (ha for 5 t wheat)Figure 3. Kilograms of nitrogen required to grow 5 t of wheat. From right: Tall,two tall cultivars of 1950 and 1960; 1960s, three semidwarfs of 1962–66; 1970s,three semidwarfs of 1971–79; and 1980s, two semidwarfs of 1981 and 1985.Source: Calculated by Waggoner (1994) from data in Ortiz-Monasterio et al. (1997).TallImproved timing and application ofnitrogen fertilizer also confer significantimprovements in uptake efficiency. Fieldexperiments indicated a marked increasein nitrogen uptake efficiency and higheryields if fertilizer was applied whenwheat plants began to pull nitrogenrapidly from the soil (about 35–45 daysafter wheat emergence in northwesternMexico) (Ortiz-Monasterio 1997). Theseimprovements in NUE result in increasedgrain protein levels and dramaticallyenhanced baking quality and nutritionalcharacteristics. Nitrogen losses into theenvironment are also significantlyreduced, along with farmers’production costs.Farmers’ adoption of this and any othernitrogen application practice will dependon the ratio of the wheat price to thenitrogen price. When this ratio becomessmaller because wheat prices fall ornitrogen prices rise, the incentive foradopting efficiency-enhancingtechnologies increases. Liberalization ofthe agricultural sector in developingcountries could be expected to increasethe adoption of technologies thatincrease nitrogen efficiency.Raised beds. The furrow-irrigated,reduced-tillage bed-planting system(FIRBS) combines the practice of plantingwheat in beds with traditional ridgetillagetechnologies. It holds immensepotential for improving irrigated wheatbasedcropping systems by making themless resource-intensive and moresustainable (K.D. Sayre, CIMMYT, pers.comm., 1997). The use of reducedtillage and crop residue managementenhances the physical, chemical, andmicrobiological properties of the soil.Planting wheat on beds provides easy6 Almost no pesticides are used to produce wheat in developing countries.

World Wheat Facts and Trends 1998/9913field access, allowing betteropportunities to apply nitrogen fertilizerwhen and where the wheat plant canuse it most efficiently. These options arenot readily available for conventionallyplanted wheat.The use of mechanical cultivationbetween the beds is of real value formore integrated weed control, especiallywhere herbicide-resistant weed speciesare prevalent. In terms of watermanagement, furrow irrigation offersimportant savings in water use and, inconjunction with the amelioratingeffects of crop residues, can improvesoils where salinization occurs. Thereduction in tillage, combined with thepresence of crop residues in irrigationfurrows, retards sediment movement inthe water, thus reducing siltation ofwaterways.Improvements in fertilizer use efficiency,weed control practices, and irrigationmanagement result in markedly betterproduction efficiencies and cost savingsfor farmers. Reduced tillage practicesfurther reduce production costs and candramatically shorten turnaround timebetween crops to ensure more timelyplanting. Estimates from trials innorthwestern Mexico place annualproduction costs for the new FIRBStechnology at nearly 30% less than forcurrent, conventional tillage practices(CIMMYT 1997a).Enhancing WheatProductivity inMarginal EnvironmentsEven if the anticipated productivity gainsin the high–potential environmentsbecome a reality, the projected growthin demand for wheat from the presentto 2020 may not be met. Gains in wheatproductivity must also be sought in thelower potential, more marginalenvironments. Low-potentialenvironments for wheat productioninclude: the semiarid areas wheredrought stress is a problem during thewheat growing season; high temperatureenvironments; and problem soil areas,especially areas with acid soils (see thebox, “Wheat Mega-EnvironmentsDefined,” p. 4). Four arguments arecommonly used to support the allocationof research resources to marginalenvironments (Byerlee and Morris 1993):• Returns to research may now behigher in marginal environments thanin more favorable environments,because the incremental productivityof further investment targeted atfavorable environments is declining.• A large number of people currentlydepend on marginal environments fortheir survival, and increasingpopulation pressure is forcing morepeople into these areas.• Because the people who live inmarginal environments are oftenamong the poorest groups of thepopulation, increased researchinvestment in these areas is justifiedon grounds of equity.• Many marginal environments arecharacterized by a fragile resourcebase. A special effort is needed forthese areas to develop appropriatetechnologies that will sustain orimprove the quality of the resourcebase over the longer term.Given these arguments, to what extentshould the research system be concernedwith technological development inmarginal environments? If theinternational wheat price continues todecline, countries that are integratedinto global trade will need to rely less ondomestic wheat production. If globalwheat prices rise with tradeliberalization, then countries will have anincentive to invest in less favorableenvironments. Given the uncertainty oflong-term price trends, the prudentstrategy for developing countries is tohave a modest level of investment inresearch for marginal environments. It isimportant to continue giving greaterweight to research investments infavorable environments, becausesignificant positive spillovers fromtechnical change in favorableenvironments benefit poor people inmarginal environments through lowerfood prices, increased employment, andhigher wages (David and Otsuka 1992).These spillovers may actually exceed thepositive benefits generated throughresearch targeted specifically at marginalenvironments (Renkow 1991).Research investments would be costeffectivein marginal areas wherespillover benefits from favorableenvironments are high. Drought-tolerantgermplasm, for example, is derived fromcrossing high-yielding wheats withwheats traditionally grown in droughtproneareas. Wheat production in WestAsia and North Africa has benefitedsubstantially from such spillover research.The international wheat research systemplays a crucial role in maximizing thespillover benefits of research onunfavorable environments acrosscountries. Maredia (1993) and Marediaand Byerlee (1999) estimated a “spillovermatrix” based on data from CIMMYT’sInternational Spring Wheat YieldNurseries (ISWYN) and demonstratedthat global research spillovers for wheat

14Global Wheat Research in a Changing Worldare large (Table 8). They found thatvarieties bred by national programs for aspecific environment possessed asignificant yield advantage in thatenvironment compared to varieties bredfor other environments. On the otherhand, CIMMYT-related wheats showed asignificant yield advantage over locallybred materials across severalenvironments, demonstrating wideadaptability and acceptability indeveloping countries.What kinds of research for marginalenvironments are likely to prove costeffective?The next sections of this reportreview research initiatives that shouldreceive priority in the effort to improvethe productivity and profitability ofwheat production in marginal areas andprotect the resource base.Improving Drought ToleranceIt is widely recognized that GreenRevolution technology disseminatedmore slowly in marginal environments,especially semiarid environmentsaffected by low water availability anddrought. Spring wheat megaenvironmentsexposed to drought—i.e.,those receiving 500 mm of rainfall orless—are MEs 4A to 4C (see the box,“Wheat Mega-Environments Defined,”p. 4). The annual gain in genetic yieldpotential in drought environments is onlyabout half of that in irrigated, optimumconditions (Rajaram et al. 1997). Twentypercent of the wheat area is affected bydrought and usually characterized byextreme poverty. In addition to thesechronically drought-prone areas, agrowing water scarcity in some irrigatedwheat production environments meansthat the wheat crop is increasinglysubject to drought caused by a reduced,less-than-optimum number of irrigations.Enhanced drought tolerance in wheatcould lead to improved food security forpopulations living in marginal areas andhelp reduce yield variability related towater scarcity in favorable environments.As noted, the search for improved wheatgermplasm that is tolerant to droughthas benefited significantly from spilloversfrom research directed at favorableenvironments. The strategy for breedingfor drought tolerance has been toTable 8. Relative yield performance of wheat cultivars of different origins in various megaenvironments(MEs), 1980– 89ME where varieties are tested a2 3 4A 4B 5A 61 High Acid Winter Winter High HighIrrigated rainfall soils rain drought temperature latitude1 Irrigated 1.00 0.95 0.84 0.90 0.88 1.02 0.942 High rainfall 0.95 1.00 0.81 0.92 0.90 0.89 0.963 Acid soils 0.89 0.96 1.00 0.85 0.90 0.98 1.004A Winter drought 0.99 0.94 0.78 1.00 0.83 0.91 0.934B Early drought 0.90 0.97 0.89 0.91 1.00 0.90 0.995A High temperature 0.88 0.86 0.92 0.82 0.89 1.00 0.926 High latitude 0.88 0.89 0.84 0.87 0.91 0.84 1.00CIMMYT/Mexico b 1.11 1.13 0.99 1.01 1.07 1.01 0.98Source: Maredia and Byerlee (1999); see also Maredia (1993) and Maredia and Eicher (1995).Note: Off-diagonal values less than one indicate that directly introduced wheat varieties from other megaenvironmentsyield less than those developed by local breeding programs in the test ME. Similarly,values greater than one (as in the case of CIMMYT varieties) indicate that directly introduced wheatvarieties from these sources yield more than those developed by local breeding programs in the test ME.a Yield expressed relative to the yield of varieties originating in that ME =1.00.b Varieties derived from CIMMYT crosses and released in Mexico.combine the high yields and inputresponsivenessof cultivars developed forfavorable environments with the droughttolerance and water-use efficiency ofgermplasm from semiarid areas.CIMMYT’s Veery lines and the advancedline Nesser are two examples.CIMMYT’s Veery lines, developedoriginally for favorable environments inthe 1970s and early 1980s, haveadapted well to less favorableenvironments, except for those withrainfall below 300 mm/yr. The Veerywheats essentially are a genetic systemthat combines high yield performance infavorable environments and adaptationto drought. The Veerys demonstrate thatefficient input use and responsiveness toimproved levels of inputs (such asavailable water) can be combined in oneplant system and used to producegermplasm for marginal (in this casesemiarid) environments. Because mostsemiarid environments differ significantlyin annual precipitation distribution, andbecause water availability also differsacross years in these environments, it isprudent to construct a genetic system inwhich plant responsiveness provides abonus whenever higher rainfall improvesthe production environment. With such asystem, improved moisture isimmediately translated into greater yieldgains for farmers (Calhoun et al. 1994;van Ginkel et al. 1998).By the mid-1980s, CIMMYT-bredgermplasm occupied 45% of thesemiarid wheat area receiving between300 and 500 mm/yr of rainfall and 1%of the area receiving less than 300 mm/yr(CIMMYT 1992), including large tracts inWest Asia and North Africa. By 1990,63% of the dryland area was planted tosemidwarf wheats (Byerlee and Moya

16Global Wheat Research in a Changing WorldAlthough wheat area and productionwere expanding in Brazil throughout the1960s and into the early 1970s, thecountry did not enjoy the benefits of the“miracle” semidwarf Mexican varietiesthat were creating a Green Revolutionelsewhere. The foremost limitation onthe adaptation of semidwarf wheats toBrazil was their extreme susceptibility totoxic levels of aluminum andmanganese, compounded by thephosphorus deficiency in the acid soilsof the region. 7Despite heavy applications of industriallime to neutralize the acidic effects ofthe soil, aluminum toxicity remained aserious problem for Mexican germplasmsown in parts of Brazil in the late 1960s.Brazilian wheat varieties tolerated theeffects of aluminum toxicity anddiseases, even though these varieties hadlow yield capacity and tall, weak straw.In the mid-1970s, CIMMYT entered intoa collaboration with Brazilian scientists tocombine Brazilian wheats’ tolerance toaluminum toxicity with the Mexicanwheats’ semidwarf stature, high yieldpotential, and wide adaptation. Morethan a decade of research yieldedsemidwarf wheats with aluminumtolerance. The new aluminum-tolerantBrazilian varieties had a yield potentialthat was 30% higher than that of theold varieties under Brazilian conditions.In addition to high yield potential, thenew Brazilian germplasm had improvedrust and mildew resistance; a betteragronomic type with regard to planttype, shorter and stronger straw, andlarger, more fertile spikes; and betterheat and drought tolerance. CIMMYTgermplasm gained longer leaf duration,7 It later became clear that soil acidity affectednot only Brazil but also many wheat-growingregions of Africa and Asia.aluminum tolerance, increasedphosphorus uptake efficiency, andresistance or tolerance to the leafspotting diseases.CIMMYT distributes outstandingadvanced lines emanating from thisshuttle project to 50 locationsworldwide. Based on data from thesemultilocational tests, outstanding linesfor yield, aluminum toxicity tolerance,and agronomic type are fed back intothe crossing program to further pyramidfavorable genes into better cultivars. Asa result, a number of high-yielding,aluminum-tolerant wheat cultivars havebeen released or recommended forrelease in several Brazilian states, andpromising cultivars have been developedfor other countries with soil acidityproblems. The cooperative shuttleprogram and distribution of the resultingmaterials has begun providing benefitsto countries in other regions, such asMadagascar, Zambia, Kenya, Tanzania,Rwanda, Cameroon, and Ecuador.Meeting the UnresolvedChallenges for MarginalEnvironmentsAt least three related factors help explainthe relatively slow rate of progress inmarginal environments compared tomore favorable areas (Morris, Belaid, andByerlee 1991). First, the climate indryland production zones severelyconstrains the yield potential of cerealcrops, so the impact of improved seedfertilizertechnologies is bound to be lessdramatic than in the more favorableenvironments where these technologiesare highly successful. Second, investmentin agricultural research targeted atrainfed areas has been modest, in partbecause such research was perceived ashaving a lower potential payoff. Third,largely because of the first two factors,many countries have been slow toimplement policies that would promotecereal production in rainfed areas, suchas policies to develop marketinfrastructure.Despite those factors, and contrary tothe common perception that theproblems of unfavorable environmentshave received inadequate attention,wheat research and cultivar developmenthave been fairly successful in theseenvironments. The principle ofmaximizing spillover benefits has workedexceptionally well in generatinggermplasm for drought-prone and acidsoil environments. Even so, some largechallenges remain to be faced: droughtstress in environments where there is norainfall during the growing period or inenvironments where rainfall levels arebelow 300 mm/yr; the need to combinedrought tolerance with heat tolerance;nutrient deficiencies (boron and zinc);and salinity. Further success in marginalenvironments depends on a substantialeffort to develop germplasm and targetit appropriately to particular settings.Although modern varieties may playsome role in boosting yields in marginalareas, germplasm will usually not be themain stimulus for rapid technical change,as is the case in favorable areas.Marginal environments would benefitmore substantially from improvements incrop and resource managementtechnologies, which often precedechanges in variety (as has alreadyhappened in Turkey) (Morris, Belaid, andByerlee 1991). For example, given thatmoisture is the primary constraint in themarginal environments, the primary

World Wheat Facts and Trends 1998/9917emphasis of technological innovationought to be moisture conservation andimprovements in moisture use efficiency.Diversification of crop and enterprisesystems is also an important means ofenhancing and/or sustaining incomes inmarginal environments, so more effortwill be needed to improve wheatproductivity in marginal environmentswhere yields remain well belowpotential. However, prospects for futuregains in productivity, as well aspromising research strategies, will differsomewhat between regions. All of thesecircumstances suggest a larger role forresearch on crop and soil managementrelative to breeding research, as well assome reorganization of research andextension strategies (Morris, Belaid, andByerlee 1991).Implications for GlobalWheat ResearchGiven that growth in population andincome will continue to spur a steadyrise in wheat demand, the challenge forsustaining wheat productivity growth isgreat. Production increases will beneeded in favorable as well as marginalproduction environments. While theneed for wheat productivity growth isgreat, it is not clear that the past highlevel of government investment inwheat research and technologydevelopment will be sustained. Here wediscuss the combination of globaltrends, research strategies, andemerging trends in research anddevelopment that will shape the globalwheat improvement system in the yearsto come and ultimately influence supplyand demand for wheat.Global TrendsThe anticipated global integration offood markets and rapid urbanizationmake it less likely that governments willgive the same emphasis to selfsufficiencyin food crop production as inthe past. Greater integration of theglobal food economy will lead tofundamental changes in food policies asdeveloping nations move away fromtraditional self-sufficiency goals andtowards increased emphasis oncomparative advantage. Some countriesmay deliberately redirect agriculturalresearch resources away from cerealcrops to higher value commercial crops,especially countries with smallerpopulations that find it feasible topurchase a significant portion oftheir food requirement on theinternational market.As the proportion of the world’spopulation living in urban areas comesto exceed the rural-based population inthe 21st century, provisioning the citieswill become the key strategic goal ofdeveloping country governments.Governments will face an increasinglydichotomous situation of rising urbandemand for cereals and decliningsupplies. Supplies will decline because ofa net movement of resources out of theagricultural sector, especially labor, andincreasing levels of diversification out ofcereal crop production into higher valuefruit and vegetable crops. Under thecircumstances, exclusive reliance ondomestic cereal crop production may notbe prudent for most governments. Inthe case of wheat, with the exception oflarge producers such as China, India,Argentina, Pakistan, Turkey, and Iran,most countries should attempt to strikea balance between domestic productionand imports to maintain urban breadprices at an affordable level. Since largeurban centers are frequently located incoastal areas, costs of transporting grainfrom overseas are often lower than thecosts of transporting grain from withinthe country.In addition to globalization andurbanization, increasing scarcity of land,labor, and water resources, along withthe removal of input subsidies, will makecereal crop production, particularlywheat and rice, less profitable at thefarm level (Pingali, Hossain, andGerpacio 1997). To sustain farmers’interest in wheat production, theresearch system needs to focus onreducing unit production costs byshifting the yield frontier and/orincreasing input use efficiencies.Research Strategies for Favoredand Marginal EnvironmentsWhile germplasm enhancement willcontinue to be the cornerstone of thestrategy for increasing productivitygrowth, innovations in crop andresource management will becomeincreasingly important for favorable andmarginal environments from aproductivity as well as from asustainability perspective. In favorableenvironments, the emphasis ought to beon enhancing input use efficiency,especially for fertilizer, water, power, andhuman labor. In the marginal, lowrainfallenvironments, the emphasis ofcrop and resource management researchought to be on moisture conservationand efficiency of water use.For favored environments, shifting theyield frontier while enhancing diseaseresistance is the priority for wheat

18Global Wheat Research in a Changing Worldgenetic improvement. Since wheatgermplasm for favorable environmentshas a high rate of spillover acrossnational boundaries, the rational strategyfor wheat improvement research wouldbe for all but the large wheat-producingcountries to rely on internationalgermplasm flows and restrict domesticdevelopment of cultivars to adaptivebreeding. Even the large wheatproducingcountries should continue touse the international germplasmexchange system as a source ofadvanced lines and diverse materials fortheir crossing programs.In the case of marginal environments,improved tolerance to physical stresseswill continue to be the priority for wheatresearch. To a large extent, productivitygrowth in the marginal environmentswill have to come from technologyspillovers from the high-potentialenvironments. Only in the case ofspecific abiotic stresses, such as droughtor high temperature, will the researchsystem have to set up a targetedresearch program. A full-scale breedingprogram for marginal wheat-growingenvironments will be profitable only incountries such as China, India, Brazil,and Argentina, which have large areasthat are marginal for wheat production.Other countries would continue to relyon the international germplasmexchange system.Emerging Trends inResearch and DevelopmentThe big breakthroughs in crop breedingresearch in the 21 st century are likely tocome from new applications of scienceto the traditional problems of shiftingthe yield frontier and enhancing yieldstability. Recent advances inbiotechnology, especially gene mappingand genetic engineering, shouldsignificantly increase the supply ofwheat germplasm with durableresistance to diseases and improvedtolerance to physical stresses. Animproved understanding of plantphysiology and crop modeling shouldalso help breeders’ efforts to shift thewheat yield frontier and developalternative crop management strategies.The application of new scientific tools toplant breeding will require intensivecollaboration between developedcountry universities, advanced researchinstitutions, international centers, andnational research systems. The spilloverbenefits from the application of newtools to crop breeding should be high,and many developing countries maychoose to benefit from spillovers ratherthan invest in full-fledged scientificcapacity to produce new productsthemselves. The size of the wheat areawill be a primary factor determiningwhether particular countries will investin biotechnology and other new wheatimprovement technology.The level of private sector participationin wheat research and development willalso have a bearing on the activities ofthe global wheat improvement system.In research and cultivar development forpure-line cereals (wheat, rye, triticale,rice, oats, and barley) and grain legumesin developing countries, the privatesector has yet to play an active role. Inresearch on hybrid cereals (maize,sorghum, and millet), where investmentsare protected by trade secrecy, theprivate sector has taken the lead ingenetic enhancement and cultivardevelopment. The role of the privatesector in hybrid maize production indeveloping countries has beenexpanding rapidly. Hybrid wheat, whenit becomes available, is expected togenerate a similar level of interest fromthe private sector. Finally, although theprivate sector plays a very considerablerole in biotechnology research, in somecountries the lack of intellectual propertyprotection could dampen the privatesector’s enthusiasm for participating incereal crop research and development.Intellectual property rights and theirimplications for the flow of geneticmaterial can also be expected totransform the way the global wheatimprovement system operates. Thewheat revolution would not haveprogressed as rapidly as it did withoutthe free and widespread globalexchange of germplasm andinformation. International spillovers ofresearch results, and the consequenteconomies of scale that resulted fromthe global flow of genetic resources,enabled wheat-growing countries bothlarge and small to benefit frominvestments in wheat research. Futuresuccess in disseminating wheattechnology worldwide depends on thecontinued uninhibited flow of geneticmaterial and information. Restrictionsimposed on such movement byintellectual property protection couldhave serious consequences ondeveloping countries’ ability to sustaingrowth in wheat productivity.

20Benefits of International Wheat Breeding ResearchAnalysis based on the actual numbers ofscientists involved in wheat improvementresearch must be treated withconsiderable caution, given the inherentconstraints of an impersonalquestionnaire and the difficulty ofenumerating scientists outside of thenational research program who conductresearch related to wheat improvement(e.g., researchers in universities). Thesefactors could lead to an underestimate.On the other hand, both the 1990 and1997 surveys asked respondents toidentify the number of full-timeequivalent scientists involved in wheatbreeding, even when they representeddisciplines other than plant breeding. Insome instances, this could lead to anoverestimate of the effort devoted towheat improvement research, asopposed to, for example, wheatcrop management.In terms of the number of scientists permillion tons of wheat production, wheatresearch intensity across the developingworld appears to be slightly greater nearthe end of the 1990s than it was earlierin the decade: 6.2 scientists per milliontons in 1997 compared to 5.3 in 1992–93 (Bohn and Byerlee 1993) (Figure 2).This difference arose largely because agreater number of wheat improvementscientists were reported for China in1997. When China is excluded, the1992–93 and 1997 figures arenearly identical.Based on these figures, wheat researchintensity may appear to be fairly stable,but it is important to note that researchby the International Food Policy ResearchInstitute (IFPRI) and the InternationalService for National AgriculturalResearch (ISNAR) has suggested thatfinancial support for agricultural researchin many NARSs has fallen in recent years.This trend has been masked at theaggregate level by continued support forresearch in strong national researchsystems such as China and India.Funding for wheat improvementresearch by NARSs appears to beincreasingly polarized, with large wheatproducingcountries continuing tosupport research while many smallercountries allocate fewer and fewerresources to national wheat research.Releases of WheatVarietiesThe national research systems ofdeveloping counties released about2,200 wheat varieties between 1966and 1997. Of these, about one-fourthwere released in 1991–97. The rate atwhich varieties are released, as measuredby the number of varieties released permillion hectares per year, seems to haveincreased in recent years in severalregions (Figure 3). Variability in rates ofrelease in some countries over time wasparticularly striking, despite the use offive-year moving averages to smooth outshort-term fluctuations.During the past 30–40 years, wheatvarieties have been released at a muchhigher rate in Latin America and sub-Saharan Africa than in the rest of thedeveloping world. Higher rates of releasemay be associated with smaller wheatareas, greater diversity in megaenvironments(that is, in the targetenvironments for wheat research; see“Wheat Mega-environments Defined,”p. 4), the rate at which diseasecomplexes change, and greaterparticipation of the private sector inwheat improvement.30252015105019921997China India Other All WANA SSA LA DevelopingAsia Asia worldFigure 2. Wheat improvement scientists per million tons of wheatproduction, developing countries and regions, 1997.Source: CIMMYT wheat impacts database.Note: WANA= West Asia/North Africa; SSA= Sub-Saharan Africa; LA= Latin America.Varieties per million hectares per year,5-year moving average25201510Sub-Saharan AfricaDeveloping countriesWest Asia/North Africa5Other AsiaIndia01965 70 75 80 85 90China95 2000Figure 3. Rate of release of wheat varieties, 1960s–1990s.Source: CIMMYT wheat impacts database.Latin America

World Wheat Facts and Trends 1998/9921Although the public sector dominateswheat improvement research indeveloping countries, there are someexceptions. Private-sector wheatimprovement research has been strongin Argentina for some time. Varietiesdeveloped by the private sector are alsosown in Brazil and Uruguay, and Chileconducted some private-sector wheatresearch in the past. In Africa, the privatesector currently appears to be importantin South Africa and Zimbabwe (Heiseyand Lantican 1998). Other Africancountries, such as Kenya and Zambia,which had no private-sector wheatresearchers in 1990, reported a modestlevel of private-sector research activity by1997. Across the developing world,however, less than 4% of the wheatarea is planted to private-sector varieties,and most of these varieties are based ongermplasm developed by thepublic sector.All of the countries surveyed have madeconsiderable use of CIMMYT wheatgermplasm. China differs from most ofthese countries, however, by using itsown material to a great extent. Theextent to which the Indian and Brazilianwheat improvement programs used theirown crosses was also notable, althougha substantial amount of the breedingmaterial in the Indian and Brazilianresearch programs was based onCIMMYT germplasm (Traxler and Pingali1998). In most other countries, theimportance of CIMMYT crosses andCIMMYT parents has not changed sincethe 1990 study.Nearly all spring bread wheatsreleased by NARSs in developingcountries are semidwarfs. The pattern ofspring bread wheat releases over time isreported in Figure 4. Of the 357 springbread wheats released by nationalresearch programs between 1991and 1997:• 56% were CIMMYT crosses,sometimes with reselection by NARSs;• 28% were NARS crosses with at leastone CIMMYT parent;• 5% were NARS crosses with CIMMYTancestry;• 8% were NARS semidwarfs withother ancestry; and• 3% were tall varieties.The percentage of spring bread wheatreleases that were CIMMYT crosses orhad at least one CIMMYT parent in1991–97 (84%) was higher than in theearlier study period, indicating that theuse of CIMMYT germplasm has notdeclined in recent years.Compared with spring bread wheats, ahigher proportion of spring durumwheats released by NARSs containedCIMMYT germplasm (Figure 5). Between1991 and 1997, of 52 spring durumwheats released by national programs:• 77% were CIMMYT crosses;• 19% were NARS crosses with at leastone CIMMYT parent;• 2% were NARS crosses with knownCIMMYT ancestry; and• 2% were tall varieties.Spring durum releases based on CIMMYTcrosses were predominant in West Asia/North Africa and Latin America.Of 106 winter/facultative breadwheats released by national programs in1991–97:• 19% were CIMMYT crosses;• 13% were NARS crosses with at leastone CIMMYT parent;• 9% were NARS crosses with knownCIMMYT ancestry;• 41% were NARS semidwarfs withother ancestry; and• 18% were tall varietiesFigure 6 presents winter/facultative breadwheat releases in the developing worldby time period. The number of releaseswas considerably higher in 1991–97 than100Tall100TallPercentage of releases80604020Other semidwarfCIMMYT ancestryCIMMYT parentCIMMYT crossPercentage of releases80604020Other semidwarfCIMMYT ancestryCIMMYT parentCIMMYT cross01966-70 71-75 76-80 81-85 86-90 91-97 66-9701966-70 71-75 76-80 81-85 86-90 91-97 66-97Figure 4. Spring bread wheat releases by time period,developing world.Source: CIMMYT wheat impacts database.Figure 5. Spring durum wheat releases by time period,developing world.Source: CIMMYT wheat impacts database.

22Benefits of International Wheat Breeding Researchin the previous study period, particularlyin West Asia/North Africa. The percentageof winter/facultative releases thatcontained CIMMYT germplasm was alsoconsiderably higher in 1991–97 thanbefore. Non-CIMMYT winter/facultativesemidwarfs were mostly Chinese releases.Adoption of ImprovedWheat VarietiesSlightly more than 80% of the wheatarea in the developing world is planted tosemidwarf varieties (Figure 7). Sixty-twopercent of the wheat area in thedeveloping world is estimated to beplanted to varieties with CIMMYTancestry. Slightly less than half of thewheat area is planted to varietiesproduced from crosses made by CIMMYTPercentage of releases1008060402001966-70 71-75 76-80 81-85 86-90 91-97 66-97or that have at least one CIMMYTparent. The proportion of wheat areaplanted to CIMMYT-related material isgreater for spring bread and springdurum wheats than for winter/facultative wheat. Table 2 summarizesthe area planted to different wheattypes in 1997.Figure 6. Winter/facultative bread wheat releases by time period, developing world.Source: CIMMYT wheat impacts database.Table 2. Area (million hectares) grown to different wheat types in 1997, classified by theorigin of the germplasmNational research system crossCIMMYT CIMMYT CIMMYT Other UnknownWheat type cross parent ancestor semidwarf Tall Landraces cultivars AllSpring bread wheat 17.8 22.4 12.6 7.7 5.2 1.4 1.0 68.1Spring durum wheat 3.4 1.2 0.02 0.11 0.3 1.5 0.1 6.7Winter/facultativebread wheat 0.6 1.9 4.2 11.6 2.2 4.1 2.6 27.2Winter/facultativedurum wheat 0.0 0.0 0.0 0.1 1.0 0.1 0.0 1.2All wheat types 21.8 25.5 16.8 19.5 8.7 7.0 3.8 103.2Source: CIMMYT wheat impacts database.TallOther semidwarfCIMMYT ancestryCIMMYT parentCIMMYT crossSpring Bread WheatSpring bread wheat is the dominant typeof wheat grown in the developingworld. Nearly 68 million hectares of thedeveloping world wheat area (includingChina) was planted to spring breadwheat in 1997. Of this area, about 60million hectares were planted tosemidwarfs, nearly 53 million hectares(88%) of which were sown to CIMMYTrelatedvarieties. CIMMYT crosses orNARS crosses with at least oneCIMMYT parent occupied about40 million hectares.The adoption of CIMMYT-related springbread wheat in 1997 is shown forregions of the developing world inFigure 8. Excluding China, 80–90% ofthe spring bread wheat area in thePercentage of all wheat area100806040200China India Other Asia WANA SSA LA DevelopingAsiacountriesUnknownLandracesTall with pedigreeOther semidwarfAny CIMMYTancestorAt least oneCIMMYT parentCIMMYT crossFigure 7. Area planted to all wheat in developing countries, 1997.Source: CIMMYT wheat impacts database.Note: WANA= West Asia/North Africa; SSA= Sub-Saharan Africa; LA= Latin America.Percentage of total spring bread wheat area100806040200China India Other Asia WANA SSA LA DevelopingAsiacountriesUnknownLandracesTall with pedigreeOther semidwarfAny CIMMYTancestorAt least oneCIMMYT parentCIMMYT crossFigure 8. Area planted to spring bread wheat in developingcountries, 1997.Source: CIMMYT wheat impacts database.Note: WANA= West Asia/North Africa; SSA= Sub-Saharan Africa; LA= Latin America.

World Wheat Facts and Trends 1998/9923developing world’s major wheat growingregions was planted to CIMMYT-relatedmaterial. The use of CIMMYT crosseswas greatest in West Asia/North Africaand Latin America, where more than50% of the spring bread wheat areawas planted to CIMMYT crosses. InChina, about one-third of the springbread wheat area was planted toCIMMYT-related germplasm, and anadditional 40% was planted tosemidwarf wheats that did not containCIMMYT germplasm.Spring Durum WheatSpring durum wheat area, which isrelatively small compared to area sownto other wheat types, is predominantlysown to semidwarfs, primarily CIMMYTrelatedvarieties. As was the case withadoption of spring bread wheats, LatinAmerica and West Asia/North Africawere the major adopters of CIMMYTcrosses in spring durum wheat. Morethan 50% of the spring durum wheatarea in West Asia/North Africa, whereover 80% of the developing world’sdurum wheat is grown, was planted toCIMMYT crosses. In Latin America, thepercentage of area planted to CIMMYTcrosses was more than 90% (Figure 9).Winter/Facultative Bread WheatIn contrast to spring bread wheat andspring durum wheat regions, areasplanted to winter/facultative breadwheat are dominated by semidwarfwheats that are unrelated to CIMMYTwheats. Among the regions wherewinter/facultative bread wheat is grown,Latin America was the only region whereCIMMYT material was dominant (Figure10). In China, where winter/facultativebread wheat occupies more than half ofthe wheat area, nearly two-thirds of thewinter/facultative bread wheat area(36% of the total wheat area) consistedof non-CIMMYT winter/facultativesemidwarfs. In the other region with alarge winter/facultative wheat area,West Asia/North Africa, nearly 40% ofthe winter/facultative wheat area wasplanted to landraces and another 35%was sown to varieties with someCIMMYT ancestry. In South Africa,which is the only country in sub-SaharanAfrica growing winter/facultative wheat,two-thirds of the area was planted totall varieties with pedigrees (versus tallvarieties whose pedigrees are unknown).LandracesRelatively little spring bread wheat arearemains in landraces. Unlike springbread wheat, large proportions of boththe spring durum wheat area and thewinter/facultative bread wheat areawere still planted to landraces in 1997.Seven million hectares of the developingworld’s wheat area were sown tolandraces and 3.8 million hectares wereplanted to unknown cultivars (i.e., theirpedigrees and origin were unknown).Landraces tended to be concentratedgeographically in West Asia/NorthAfrica. Landraces covered slightly lessthan 20% of the spring durum area inWest Asia/North Africa. In Ethiopia, theonly country in sub-Saharan Africawhere durum wheat is planted,landraces covered nearly 80% of thewheat area. As noted, landracesoccupied nearly 40% of the winter/facultative wheat area in West Asia/North Africa.CIMMYT/NARS CollaborationThe data for numbers of releasedvarieties and area planted indicate thatCIMMYT plays a major role in wheatimprovement research for developingPercentage of total spring durum wheat area100UnknownPercentage of total facultative/winter bread wheat area100Unknown80Landraces80Landraces60Tall with pedigreeOther semidwarf60Tall with pedigreeOther semidwarf40200China India Other Asia WANA SSA LA DevelopingAsiacountriesAny CIMMYTancestorAt least oneCIMMYT parentCIMMYT crossFigure 9. Area planted to spring durum wheat in developingcountries, 1997.Source: CIMMYT wheat impacts database.Note: WANA= West Asia/North Africa; SSA= Sub-Saharan Africa; LA= Latin America.40200China India Other Asia WANA SSA LA DevelopingAsiacountriesAny CIMMYTancestorAt least oneCIMMYT parentCIMMYT crossFigure 10. Area planted to facultative/winter bread wheat indeveloping countries, 1997.Source: CIMMYT wheat impacts database.Note: WANA= West Asia/North Africa; SSA= Sub-Saharan Africa; LA= Latin America.

24Benefits of International Wheat Breeding Researchcountries. It is important to recall thatthe Center’s contribution is not theresult of independent activity;international wheat improvementresearch is collaborative and progressdepends on international testing by anetwork formed by CIMMYT andnational research systems worldwide(Maredia and Byerlee 1999).Furthermore, every cross is the result ofa conscious decision by the breeder andinstitution that made the cross, ofdecisions made earlier by other breeders,and ultimately of breeding decisionsmade by the farmers who tendedlandraces over the centuries.Trends and ObservationsThe extensive data collected for the1990 and 1997 studies allow us to lookfor longitudinal trends, albeit using arather short time frame. Our estimatesof wheat area have relied on fairlysimple ways of assessing the CIMMYTcontribution. Here, we recapitulate someof those measures and present anadditional one based on a geometricrule developed by Pardey et al. (1996).This additional measure analyzes avariety’s pedigree by applyinggeometrically declining weights to eachlevel of crossing for as many generationsas desired. The weights applied to theearliest generation included in theanalysis are increased to make the totalof all weights sum to 1. 2For comparison, we present estimatesfor 1990 based on the data analyzed byByerlee and Moya (1993). In presentingthe 1997 data, we provide figures thatboth include and exclude China,because only a few spring bread wheatzones in China were covered in the1990 study and because China does notuse CIMMYT germplasm as extensivelyas other developing countries.Calculations for spring bread wheat arepresented in Figure 11 and for winter/facultative bread wheat in Figure 12. 3Excluding China, spring bread wheatarea planted to CIMMYT crossesdeclined between 1990 and 1997.During the same time, however, areaplanted to NARS crosses with CIMMYTparents increased, as did the areaplanted to spring bread wheat with anyCIMMYT ancestry. The decline in springbread wheat area planted to CIMMYTPercentage10019901997 without China801997 with China6040200CIMMYT CIMMYT Any Geometriccross cross or ancestor ruleCIMMYT parentFigure 11. CIMMYT contributions to springbread wheat planted in developingcountries, 1997.Source: CIMMYT wheat impacts database.crosses can be explained by somewhatlower planting of direct CIMMYT crossesin India, Turkey, and Pakistan—threelarge developing-country wheatproducers. By the geometric rule, in1990, approximately 45% of the geneticcontribution to spring bread wheatcould be attributed to CIMMYT. In the1997 data, this figure fell to slightlymore than 40%, because of the declinein area planted to CIMMYT crosses (ascrosses are given the most weight in thegeometric index). 4 As expected, whenChina is included the CIMMYTcontribution declines by all measures;the decline is proportionately the lowestwhen using the “any ancestor” rule(Figure 11). The reason for this finding isthat, compared to other breedingprograms, the Chinese wheat programoften uses CIMMYT material at anearlier stage of the crossing process.Percentage10019901997 without China801997 with China6040200CIMMYT CIMMYT Any Geometriccross cross or ancestor ruleCIMMYT parentFigure 12. CIMMYT contributions to winter/facultative bread wheat planted indeveloping countries, 1997.Source: CIMMYT wheat impacts database.2 For example, if the analysis were carried back to the level of great-grandparents, the source of the final cross would be given a weight of 1/2, the source ofeach of the parents would be given a weight of 1/8, and the source of each of the grandparents would be given a weight of 1/32. The next fraction in thisseries is 1/128, but the source of each of the great-grandparents would be given a weight of 1/64 to ensure that the weights sum to 1.3 Calculations for spring durum wheat are being revised. As a result, estimates for spring durum wheat and for “all wheat” are not presented; however thepatterns for spring durum and all wheat are fairly clear.4 Although the estimates for spring durum wheat are being recalculated, it is likely that in comparison with spring bread wheat, CIMMYT crosses are moreimportant in terms of area planted in spring durum than in spring bread wheat; NARS crosses with at least one CIMMYT parent, and NARS crosses with earlierCIMMYT ancestry, are less important in durum wheat. Genetic contribution as measured by the geometric rule is greater in spring durum wheat than in springbread wheat, again because of the higher weight given to the final cross. Slightly more than 50% of the genetic contribution to spring durum wheat plantedin developing countries would be attributed to CIMMYT by this rule.

World Wheat Facts and Trends 1998/9925As expected, the data on releases andarea planted show that CIMMYT hasmade a smaller contribution to winter/facultative wheat breeding in developingcountries relative to its contributions tospring bread or spring durum wheatbreeding. Even so, CIMMYT’scontribution to winter/facultative wheathas grown substantially since 1990. NoCIMMYT winter/facultative crosses wereplanted in 1990; in 1997, a small areawas planted to such crosses. ExcludingChina, the proportion of winter/facultative wheat planted to varietieswith some CIMMYT ancestry tripledbetween 1990 and 1997 (Figure 12).Within China in 1997, a little more than10% of the winter/facultative wheatarea was planted to a variety with someCIMMYT ancestry.The final figures for all wheats takentogether are not yet available, but theywill most likely parallel the estimates forspring bread wheat, because that is thedominant wheat type in the developingworld. It then follows that the areaplanted to CIMMYT crosses is likely tohave declined somewhat between 1990and 1997, while the area planted toNARS varieties with at least one CIMMYTparent and to NARS varieties with earlierCIMMYT ancestry increased. By thegeometric rule, it appears that in 1990and 1997 CIMMYT accounted for justunder 40% of the genetic contributionto all wheat planted in the developingworld (excluding China). If China isincluded, the 1997 CIMMYT contributionwould probably be less than 30%.Outstanding IssuesIn assessing the impacts of internationalwheat breeding research in thedeveloping world, several issues meritfurther study and analysis. These includethe rate at which varieties are replaced infarmers’ fields, the rates of genetic gainin wheat yield and yield gain in farmers'fields, and the future of internationalcollaboration in wheat improvementresearch.Varietal ReplacementOur most recent data indicate that, asreported in the 1990 study by Byerleeand Moya (1993), a significantproportion of the total wheat area is stillplanted to older improved varieties.Despite the fact that farmers indeveloping countries have widelyadopted improved varieties, the rate atwhich older improved seed is replaced byseed of newer improved varieties remainsunacceptably slow. Farmers benefitneither from the improved yield potentialof newer varieties nor from their superiordisease resistance.One measure of the rate at whichvarieties are being replaced is the age ofvarieties in farmers’ fields, measured inyears since release and weighted by thearea planted to each variety (BrennanTable 3. Weighted average age (years) of improved varieties in farmers’ fields, 1997CountryAge of varietiesAfghanistan, Zimbabwe 14and Byerlee 1991). Based on thisindicator for 1990 and 1997, varietalreplacement had become more rapid by1997 in only 12 of the 31 countries forwhich comparisons could be made.Table 3 presents a classification ofcountries by weighted average age ofimproved varieties in farmers’ fields in1997. Note that only improved varieties(semidwarfs and improved tall varieties)are included in these calculations.Among developing countries, only two,Zimbabwe and Afghanistan, had anaverage age of varieties in farmers’ fieldswithin less than six years. This length oftime is important because it is the periodestimated (based on weighted averages)in which rust resistance derived from asingle resistance gene can be maintained(Kilpatrick 1975). (Rust is the mostimportant disease of wheat worldwide;for a discussion of the importance of rustresistance, see “CIMMYT’s strategy toachieve durable rust resistance inwheat,” p. 10.) In Zimbabwe, it appearsthat the private sector’s involvement inwheat research may have played a role inthe rapid turnover of wheat varieties. InAfghanistan, external aid following theRussian withdrawal included widespreaddistribution of wheat seed.Most Latin American countries, with theexception of Peru and Mexico, seem toreplace their varieties in the field morerapidly than other developing countries,which is consistent with earlier findings(Byerlee and Moya 1993). In Mexico,however, varietal replacement was muchmore rapid in the past (Brennan andByerlee 1991), primarily because of ashift in the major wheat-growing areasof northwestern Mexico from breadwheat to older durum wheat varieties.Source: CIMMYT wheat impacts database.

26Benefits of International Wheat Breeding ResearchIn contrast to Latin America, in mostnations of West Asia/North Africa theweighted average ages of varieties in thefield exceeded 14 years. Interestingly,even some large wheat-producingcountries such as India had weightedvarietal ages exceeding 12 years,although wheat varieties were replacedmuch more rapidly in some regions ofIndia, particularly the northwest (Byerleeand Moya 1993). Factors affecting therate of varietal replacement in wheat arediscussed from a theoretical perspectiveby Heisey and Brennan (1991) andempirically by Heisey (1990) and byMwangi and colleagues (see, forexample, Alemu Hailye et al. 1998;Regassa Ensermu et al. 1998; and HailuBeyene, Verkuijl, and Mwangi 1998).Genetic Gains in Wheat Yields andYields in Farmers’ FieldsConsiderable debate has surrounded thequestion of whether breeders arecontinuing to make genetic gains in yieldpotential in the major cereals or whethermost recent progress has come fromraising the bottom of the yielddistribution by increasing resistance tostress (see Evans and Fischer,forthcoming; Mann 1999). It appearsthat the genetic yield potential of wheathas continued to increase (Evans andFischer, forthcoming; Sayre, Rajaram andFischer 1997). An extensive review ofstudies about gains in wheat yields indeveloped and developing countries alsoconcluded that there is no convincingevidence that genetic gains in wheatyield potential have leveled off (Rejesus,Heisey, and Smale 1999). As isapparently the case in rice and maize, inwheat the larger proportion of geneticgains in yield also results from increasedstress tolerance rather than gains in yieldpotential per se. Much of the progress indeveloping stress tolerance in wheat hascome from dramatically improvedresistance to wheat diseases, particularlythe rusts (Sayre et al. 1998).Meanwhile, there is ample evidence thatyield advances in farmers’ wheat fieldshave slowed. Worldwide, some of thisslowdown may be partially explained byreduced growth in demand for wheat,but it is noteworthy that yield increasesin advanced wheat-producing areas ofdeveloping countries (e.g., northwesternMexico and the Punjabs of India andPakistan) have also slowed in the past10–15 years. The reasons for slowergrowth in yield gains in farmers’ fieldsare many and complex; it is quite likelythat crop management issues andresource degradation play importantroles (see Part I of this report for adiscussion of the opportunities andconstraints related to improving wheatproductivity).The Future of InternationalCollaboration in WheatImprovement ResearchDuring the 1990s, the framework anddynamics of crop improvement researchentered a period of rapid change.Increased private-sector investment incrop improvement, changes inintellectual property rights regimes, andpotential technical changes cametogether to alter the ways that breedersand farmers can use seed. Thesedevelopments continue to transform thedynamics of germplasm exchange, insome cases limiting the free exchange ofgermplasm on which the success of theCIMMYT/NARS international wheatbreeding effort was built. Althoughmany of these changes are taking placeprimarily in industrialized countries, it isclear that the impacts will be global.During this same period, researchintensity for wheat breeding in thedeveloping world seems to haveincreased, as has the rate at whichwheat varieties are released. Despitethese indicators of progress, ancillaryevidence on the funding andorganization of wheat research in all buta few of the largest NARSs and inCIMMYT’s own wheat program suggeststhat the international system for wheatimprovement research faces continuingchallenges in the years ahead.The structure and organization of theinternational wheat breeding effort wasdiscussed in Part I of this report and hasbeen extensively analyzed by Marediaand Byerlee (1999), Byerlee and Traxler(1996), and Traxler and Pingali (1998). Itis clear from these analyses and from thedata presented here that the CIMMYT/NARS collaborative effort in wheatimprovement could remain central toproducing significant benefits for wheatproducingnations and farmers well intothe next century. Whether the politicalwill and financial resources to sustainthat effort will be available, however,remains an open question.

Part 3The World Wheat Economy, Post-2000:Focus on AsiaPrabhu L. PingaliWorld Wheat Facts and Trends 1998/9927Recent projections by the InternationalFood Policy research Institute (IFPRI)indicate that, by 2020, two-thirds of theworld’s wheat consumption will occur indeveloping countries, where wheatimports are estimated to double by2020. As noted earlier in this report,wheat demand worldwide is calculatedto rise by 40% from 1993 to 2020 toreach 775 million tons. The expectedincrease in demand is partly motivatedby population growth but also resultsfrom substitution out of rice and coarsegrain cereals as incomes rise andpopulations become increasingly basedin urban areas.When countries move from low- tomiddle-income status, per capitaconsumption of maize and rice for fooddeclines, while per capita consumptionof wheat tends to rise (see “Wheat forAsia’s Increasingly Westernized Diets,” p.28). 1 In China for example, wheatconsumption is expected to rise from 83kg per capita per annum in 1993 to 88kg per capita per annum in 2020(Rosegrant et al. 1997). India, on theother hand, is expected to see per capitawheat consumption rise from 55 to 64kg per annum.Competitiveness ofWheat Production inDeveloping CountriesAlthough wheat demand is expected torise, it is not clear that expandingdomestic production will be competitive1 At very high income levels, per capita wheatconsumption also tends to many developing countries. Thedeveloping world is undergoingunprecedented policy reforms that couldsignificantly affect growth in wheatproductivity over the next severaldecades and significantly influence thelevel of wheat imports. The reforms fallinto two broad categories: liberalizationof the agricultural sector and increasedglobal economic integration. As Part I ofthis report indicated, the primary impactof these reforms is to move developingcountry agriculture away from itstraditional focus on food self-sufficiency.Crop choice and land use decisions willbe made increasingly on the principles ofcomparative advantage rather than onthe imperatives of domestic food needs.The nonagricultural sector will competemore strongly for production inputs(even those available on the farm, suchas family labor), and these inputs will bevalued at their true opportunity cost.Liberalization of the agricultural sectorcould imply an almost total removal ofpolicy protection and support to thecereal crops sector. Output pricesupports, input subsidies, andpreferential access to credit will all begradually phased out or at leastsubstantially reduced. Infrastructural andresearch support for the cereals sectorcan also be expected to decline asgovernments diversify their agriculturalportfolios to include crops that arecompetitive on the global market.The anticipated liberalization ofdeveloping country economies and theirincreased global integration will havesignificant consequences for theorganization and management ofagricultural production. The anticipatedwithdrawal of labor from theagricultural sector will lead to anincrease in the opportunity cost of laborand make smallholders’ intensive cerealproduction systems less profitablerelative to other income-earning andlivelihood opportunities. Land and waterresources will face similar competitivepressures from the nonagriculturalsectors. Provisioning food for thegrowing urban conglomerates isexpected to be a major challenge of the21 st century, but it is important torecognize that domestic sources ofsupply will have to compete with oftencheaper international sources of supply,especially in coastal mega-cities.Even with the anticipated success inenhancing growth in wheat productivity,the quantity of wheat imported into thedeveloping world is expected toincrease. Traditional wheat exporters,such as Argentina, Australia, Canada,and the US would definitely benefitfrom the increased global demand, andnew exporters are expected to emergein Latin America and Africa. Countrieswith high land-to-labor ratios, goodmarket infrastructure, and suitableagroclimatic conditions are allcandidates to expand wheat exports.Brazil and Mexico are two countries towatch in this regard. Mexico is alreadyemerging as an important player in theexport market for durum wheat.

28The World Wheat Economy, Post-2000Special Report:Wheat for Asia’s Increasingly Westernized DietsPrabhu L. Pingali and Mark W. RosegrantThe increasing Westernization ofAsian diets (in other words, thesubstitution out of rice andincreased consumption of breadand high-value foods) is aninevitable consequence of risingincomes, urbanization, andeconomic modernization. Thegrowth in per capita wheatconsumption is an excellentindicator of the extent of dietdiversification in Asia. Asian wheatdemand in 2020 is projected to bearound 322 million tons ascompared to 205 million tons in1993. The current economicdownturn in Asia is anticipated tohave only a minimal effect on theextent of diet diversification andper capita wheat demand.Most countries of East andSoutheast Asia can meet theirincreasing demand for wheat onlyby importing more wheat. Chinaand the countries of South Asia,however, will not find iteconomically or politicallyexpedient to rely exclusively onimports for meeting theiradditional wheat requirements. Inthese densely populated countries,domestic wheat production willhave to increase dramatically toensure reliable wheat supplies andfood security.Changing DietaryPatterns and WheatDemandAcross Asia, rapid economic growthand urbanization are creatingdramatic changes in dietarypatterns. As incomes rise, ricebecomes an increasingly inferiorgood in Asia. Households tend toincrease their consumption of breadand high-value food such as meat,poultry products, fruit, andvegetables.The income elasticity of demand forwheat rises rapidly with incomegrowth as countries move from lowto middle income levels; at middleto high income levels, incomeelasticity of demand remainsrelatively stable; and at very highincome levels, it turns negative. Eastand Southeast Asian countries fallinto all three of these categories,although Japan, with a per capitawheat consumption of 50 kg peryear, is the only Asian countryexhibiting negative incomeelasticity of demand for wheat.The IFPRI IMPACT model (Rosegrantet. al. 1997) projects that Asia willaccount for 42% of global wheatdemand in 2020 as opposed to 37%in 1993. China, with an anticipateddemand in 2020 of 152 million tons,and India, with an anticipateddemand of 96 million tons,together will account for 77% ofAsian wheat demand. According tobaseline estimates, Asian demandfor wheat is anticipated to grow atleast 3% per year through 2020.Even assuming that productivitywill grow steadily in Asia’s wheatgrowingareas, large importvolumes are projected (see below).The current economic slowdown inAsia should have only a modesteffect on wheat demand, which isanticipated to be 5% lower thanbaseline estimates. Reduced growthscenarios were estimated using a50% reduction in growth innonagricultural GDP, beginning in1999, for all Asian developingcountries. If the economicdownturn were to affect onlySoutheast Asia and South Korea,the impact on wheat demandwould be even more modest.Demand for wheat in 2020 inSoutheast Asia and Korea wouldfall an estimated 800,000 t and300,000 t, respectively, relative to2020 baseline projections.The countries of Southeast Asia areexperiencing the most rapidgrowth in wheat consumption astheir incomes rise; it is anticipatedthat they will have positive andrising income elasticities of demand

World Wheat Facts and Trends 1998/9929through the 2020 time horizon. Percapita wheat consumption in theregion, which was less than 5 kg/yrin 1961, increased three-fold over a30-year period, reaching 15 kg percapita per annum by 1993. It isexpected to double, relative to its1993 level, by 2020, despite theeconomic crisis of 1997.Wheat demand patterns in SouthAsia are different from those ofEast and Southeast Asia. Wheat isthe traditional cereal in northernIndia, Pakistan, Nepal, and northernparts of Bangladesh. Per capitaconsumption in these areas tends tobe relatively stable with respect toincome growth; per capita wheatconsumption is projected at around2.8% per annum through 2020.Increasing demand in South Asiawill continue to be driven primarilyby population growth. Incomeinduceddiversification in diets andthe substitution out of rice areobserved mainly in southern andeastern India, parts of Bangladesh,and Sri Lanka. Per capita growth inwheat consumption in these areas isprojected to be approximately 3.4%per annum through 2020.Sources of WheatSupplyAsia’s increased demand for wheatwill have to be met through acombination of increased importsand enhanced domesticproduction, where technologicallyfeasible. The technologicalpotential for increasing wheatproduction is limited in thecountries of East Asia (exceptChina) and Southeast Asia, whereaccelerating demand for wheat willbe met primarily throughexpanding imports. In China, India,Pakistan, Bangladesh, and Nepal,wheat demand will be met thoughmarginal imports and expandeddomestic supplies.Asian wheat imports are expectedto escalate rapidly over the nexttwo decades relative to the 1996level of 30 million tons. IFPRIprojects Asian wheat imports willreach 62 million tons by 2010 and75 million tons by 2020. Asia’scurrent economic downturn isanticipated to have only a modestimpact on wheat imports, a drop offive million tons at the most,relative to IFPRI’s baselineprojection. If China remainsunaffected by the Asian economiccrisis, the drop in import demandwill be a mere one million tons.Despite anticipated growth indomestic supplies, China isexpected to import approximately22 million tons of wheat by 2020.Wheat imports in 2020 to East andSoutheast Asia (excluding China)are projected at 24 million tons, atwo-fold increase over 1993 levels.For China as well as India and otherSouth Asian countries, rapidgrowth in domestic wheatproduction is absolutely crucial tomeeting growing requirements forwheat. Wheat demand in Chinaand South Asia in 2020 will reach300 million tons, but there areserious concerns regarding theprospects for sufficiently increasingdomestic wheat production. Theseconcerns relate to technologicalopportunities for productivitygrowth in the favorable andmarginal wheat growingenvironments, to policy issues, andto farm-level incentives forincreasing productivity, especiallywith the increased globalintegration of food markets.

30The World Wheat Economy, Post-2000Special Report:Wheat in Kazakhstan—Changing Competitivenessand Sources of Productivity GrowthJim Longmire and Altynbeck MoldashevKazakhstan faces many challengesin its transition from a centrallyplanned economic system towardsa more market-oriented one. Likemany other former Soviet republicsin transition, Kazakhstanexperienced a sharp contraction inits economy and high inflation inthe years immediately followingindependence. The overalleconomic situation has sinceimproved slightly, but theagricultural sector remainsparticularly affected by the politicaland economic changes.One of Kazakhstan’s primaryresponsibilities in the Sovieteconomy was the production andexport of wheat. Wheat is still thecountry’s principal crop, andKazakhstan remains the dominantwheat producer of Central Asia aswell as the third largest producer ofthe former Soviet republics(following Russia and the Ukraine).Kazakhstan is also the mostimportant producer of high-proteinwheat in Asia and Europe.Wheat production takes place intwo predominant environments,the dryland steppes of the northand the irrigated and rainfedsouthern environment. Thenorthern area is planted to springwheat varieties owing to itsextreme winter conditions andprecipitation patterns. Conditions inthe southern regions of the countryare suitable for producing winterwheat. Spring wheat makes up theoverwhelming majority of wheatcultivated in Kazakhstan.A combination of factors, includingmacroeconomic policies and policieswithin the agricultural sector, haveprecipitated large adjustments inKazakhstan’s wheat economy. Thegovernment has drastically reducedits investment in the agriculturalsector in favor of increased activityin energy and other extractivesectors, where relatively largerMillion hectares/million tons20161284Production01960 70 80 86 88 90 92 94 96 98Figure 1. Kazakhstan wheat area andproduction, 1960–98.Areapotential exists for foreign exchangeearnings and foreign investmentopportunities. Decreases in fundingand personnel have placed severeconstraints on the breadth andscope of research and extensionactivities that can be carried out.The government has also taken alaissez faire attitude in price policiesfor both agricultural inputs andoutputs.In response to the existingincentives, dramatic decreases inwheat area, output, and yields haveoccurred in the last decade (Figures1 and 2). Wheat area fell from morethan 15 million hectares in the late1980s to slightly over 9.5 millionhectares in 1998. Yields have alsodeclined in the last decade,Tons/hectare1. 70 80 86 88 90 92 94 96 98Figure 2. Kazakhstan average wheat yield,1960–98.Source: Kazakhstan State Committee for Statistics,FAOSTAT.

World Wheat Facts and Trends 1998/9931dropping from an average ofslightly over 1 t/ha between 1985and 1990 to under 0.7 t/ha between1993 and 1998. The combineddecline in area and yield causedwheat production to fall from anaverage of 14.5 million tons to lessthan 5 million tons during the sametime period. Farm profitability andincomes have fallen as well, and thelevel of input use on farm has beenconsiderably reduced (see table).Phosphate use was 315,000 t in1992 and 55,000 t in 1996. Nitrogenuse shows a similar trend, fallingfrom 150,000 t to 63,000 t between1992 and 1996. Many transactionsfor inputs and services nowcommonly rely on bartering.However, the lack of investmentand the decline in public researchexpenditures have not been theonly causes of reduced productivityin the wheat sector. Decades ofunsustainable cropping patternsTable 1. Fertilizer use (000 t) in KazakhstanInput 1992 1993 1994 1995 1996Nitrogen 150 86 65 64 63Phosphate 315 231 50 25 55Potash 10 7 6 6 6Source: FAOSTAT (1998); World Bank (1999).have contributed to degradation ofthe resource base. Some of themost inappropriate land for wheatcultivation has already beendropped from production.What are the potential sources offuture wheat productivity growth?The inherent challenge of dealingwith the variability resulting fromclimate-induced croppingconditions, particularly in northernKazakhstan, will always be present.Kazakhstan is fortunate to have astrong tradition of agriculturalresearch. Continued research inwheat improvement and relatedseed management should lead togreater productivity. Gains fromimproved disease resistance alonemay be substantial. Anotherpromising area of research fallswithin the realm of agronomy andcrop management improvement.Practices such as more timelyplanting, better moisturemanagement, better weed control,and improved crop rotation haveall successfully increased yields infield trials and could possiblyprovide farmers with a favorablecost/benefit tradeoff.The present and potentialtechnological possibilities arenumerous, but they must beexamined in the context of theenvironment in which farmersoperate. The current disincentive toinvest in inputs, machinery, andnew technologies has majorimplications for agriculturalresearch and extension. The inputintensivetechnologies developedduring the Soviet era are generallyno longer competitive under thedeveloping market-based system. Inthe next few years at least, farmersare likely to adopt only thosemethods that will incur minimalcosts. Further research in targetingthe most useful and cost-efficienttechnologies at the farm level isneeded. Changes, albeit gradual, infarm organization and ownership,will also necessitate adjustments incurrent extension methods.Note: For more detail on wheat inKazakhstan, see Longmire andMoldashev (1999).

32The World Wheat Economy, Post-2000Targeting WheatResearch andDevelopmentInvestmentsWith the progression towards globalintegration, the competitiveness ofdomestic wheat production can bemaintained only through dramaticreductions in the cost per ton ofproduction. As discussed in Part I of thisreport, some of the high pay-offstrategies for increasing thecompetitiveness of wheat, particularly inthe high-potential environments,include shifting the yield frontier,increasing yield stability, and enhancinginput use efficiencies. Research anddevelopment (R&D) in the lowerpotential environments ought toconcentrate on improving yield stabilitythrough the development of geneticmaterials and production systems withimproved tolerance to drought andother physical stresses.The returns to cereal crop R&Dinvestments will not be uniformly highin all countries and all environments. Toensure adequate future grain supplies,it will be crucial to target theseinvestments carefully. Some of thefactors that ought to be taken intoaccount in determining the returns toR&D for a particular crop are the size ofthe domestic market, export potential,and the proportion of high-potentialarea under the crop.The size of the domestic market isdetermined by aggregate populationprojections as well as by the prospectsfor income growth. Countries with largepopulations, such as India and China,would want to invest in improvingdomestic cereal supplies to buffer theconsumer against the vagaries of theinternational market. The demand forproductivity-enhancing technology isgenerally expected to be the greatest inthe high-potential productionenvironments, especially in countrieswith a large domestic demand forcereals. In the case of wheat, theirrigated environments were the primarybeneficiaries of the Green Revolution,and in the short to medium term,sustaining productivity growth in theseenvironments depends largely onsustained R&D investments.Given the above, where should oneexpect to see high returns to investmentin wheat R&D? China and India are theleading countries for anticipated highreturns to investment in wheat R&D,primarily directed towards meetingdomestic demand. Mexico, South Africa,and Egypt are also likely to find wheatresearch investments directed to thedomestic market to be attractive (Egyptin particular through spillover benefitsgained from other regions of the world).Among wheat exporting countries, thelargest gains through such investments,specifically for spring wheats, are likelyto be in Australia and Argentina. Theopen question in the case of wheat isthe potential of the countries in theformer Soviet Union to respond toglobal wheat demands throughincreased exports (see “Wheat inKazakhstan: Changing Competitivenessand Sources of Productivity Growth”).The foregoing discussion is not meant toimply that countries with smaller wheatgrowing areas will find their productionunprofitable or that they will be unableto benefit from improved technologies.By making modest investments andmaximizing spillover benefits fromother countries with similaragroclimatic conditions, several ofthese countries may experienceproductivity gains similar to those inlarger wheat producing countries.Similar caution ought to be expressedin making R&D investments formarginal environments. Wherespillovers from favorable environmentsare possible, modern varieties do moveinto marginal environments underappropriate market conditions andgenerally perform better thantraditional varieties. Where spilloversare not possible, and where a shift inthe yield frontier is potentially the onlysource of productivity growth, publicresearch investment will be needed todevelop varieties with appropriatetolerance to physical stresses.The importance of the marginalenvironment to the domestic cerealsector and the availability of alternativesources of livelihood for people living inthat environment are majordeterminants of the level of publicresearch investment for improving cropproductivity. In Sub-Saharan Africa andSouth Asia, continued high levels ofinvestment in marginal environmentsare imperative to ensure long-termfood security for people living in thoseareas, in the absence of otherlivelihood opportunities. Finally,investments in marginal environmentscannot be based solely on efficiencycriteria: equity considerations willcontinue to play a major role ininvestment decisions.

Part 4Selected Wheat StatisticsPedro Aquino, Federico Carrión, and Ricardo CalvoWorld Wheat Facts and Trends 1998/9933The following tables present statisticsrelated to wheat production, trade,utilization, experimental yield, type ofwheat, prices, and input use. Thesestatistics reflect the latest informationavailable at the time of publication.Countries are classified as either“developing” or “high income” basedon the criteria used by the World Bank inits World Development Indicators (1999).Countries classified as “developing” hada per capita GNP lower than US$ 9,655in 1997, whereas high-income countrieshad a per capita GNP exceedingUS$ 9,656. Countries in Eastern Europeand the former Soviet Union (FSU) aretreated separately. Traditionally includedas “developed” countries in FAOstatistics, most of these countries wouldbe classified as developing countries byWorld Bank criteria.The first two sets of tables presentproduction statistics and consumptionstatistics. Developing countries and thosein Eastern Europe and the FSU areincluded in the individual countrystatistics if they consumed (or produced)at least 100,000 tons of wheat per year.Developing countries are classified as“wheat producers” if they producedmore than 100,000 tons of wheat peryear, regardless of import andconsumption levels. Developingcountries that produced less than100,000 t/yr, but that produced at least50% of their total wheat consumption,are also classified as producers. Otherdeveloping countries that consumedover 100,000 t/yr are defined as “wheatconsumers.” High-income countries areclassified in the same way, usingminimum levels of production orconsumption of 1 million tons. A threeyearaverage of the latest data availablewas used in the classification.Unless otherwise indicated, the regionalaggregates include data from all thecountries in a particular region, includingthose countries for which data have notbeen reported individually. For a list ofcountries belonging to each region, seeAppendix A. Regional means areappropriately weighted; thus they maynot exactly equal the mean of theaverage values presented for eachcountry. Former Czechoslavakia, formerYugoslavia, and the FSU were dividedinto separate countries, for whichstatistics are reported individually.Notes on the VariablesThe data source for all production andconsumption statistics is FAO, FAOSTAT(1999).Growth rates were calculated using thelog-linear regression model:In Y = α + ßt + ε,where In Y is the natural logarithm of Y,t is time period (year), α is a constant, ßis the growth rate of Y, and ε is the errorterm. The function describes a variable Y,which displays a constant proportionalrate of growth (ß>0) or decay (ß

34Selected Wheat StatisticsProduction statisticsAverage wheat area, yield, and production, 1995-97Growth of wheat area (%/yr)REGION / COUNTRY Harvested area (000 ha) Yield (t/ha) Production (000 t) 1951-66 1966-77 1977-85 1985-97 *Eastern and Southern Africa 2882 1.6 4709 2.1 1.0 0.4 -1.2Ethiopia 877 1.3 1113 1.0 -5.7 4.0 10.1Kenya 159 1.9 295 -0.9 -1.1 -0.4 1.4South Africa 1350 1.7 2328 3.1 3.4 0.9 -4.9Sudan 302 1.8 539 8.4 13.4 -17.9 12.5Zambia 18 3.2 57 n.a. 27.6 13.5 12.4Zimbabwe 50 4.1 204 4.6 19.8 -7.5 0.9North Africa 6037 2.0 11767 0.0 0.9 -1.1 1.3Algeria 1595 1.1 1715 0.2 1.5 -3.3 -1.1Egypt 1039 5.6 5769 -1.8 0.1 -1.1 6.8Morocco 2558 1.2 3108 0.7 -0.2 0.4 1.8Tunisia 827 1.4 1144 -1.5 2.8 -0.3 1.2West Asia 21436 1.8 38220 2.5 1.3 -0.1 0.7Afghanistan 2025 1.3 2620 1.3 1.1 -4.2 1.7Iran 6398 1.6 10429 5.0 1.1 1.6 0.3Iraq 1480 0.8 1200 0.7 -1.0 -1.6 4.0Saudi Arabia 326 4.8 1549 6.6 -1.4 29.0 -5.1Syria 1675 2.3 3765 -0.1 5.2 -3.6 3.6Turkey 9363 2.0 18393 2.6 1.3 -0.3 0.2Yemen 103 1.5 150 6.0 9.8 -1.6 3.3South Asia 35240 2.4 85021 2.1 3.5 1.9 1.0Bangladesh 683 2.0 1356 3.5 7.9 17.3 1.2India 25585 2.6 65713 2.3 4.3 1.6 1.0Myanmar 98 0.9 85 14.5 -4.0 4.9 -1.1Nepal 648 1.6 1009 -1.1 8.8 4.2 2.5Pakistan 8218 2.1 16853 1.7 1.1 2.0 1.1East Asia 29921 3.8 112378 -0.1 1.3 0.3 0.1China** 29510 3.8 112024 -0.3 1.4 0.2 0.1Mongolia 328 0.7 237 24.7 -0.7 2.9 -3.4North Korea 80 1.4 108 8.7 -6.3 0.0 -0.8Mexico, Central America, and Caribbean 850 4.2 3525 1.0 0.0 6.4 -3.0Mexico 837 4.2 3500 1.1 -0.2 7.0 -2.9Andean Region, South America 309 1.1 341 -0.2 -2.7 -0.9 1.4Bolivia 140 0.9 122 6.5 2.5 0.8 3.9Peru 109 1.2 132 -0.5 -1.9 -3.3 0.8Southern Cone, South America 8146 2.2 18160 0.3 2.1 1.1 -1.9Argentina 5893 2.3 13386 1.1 -1.3 5.2 1.2Brazil 1440 1.7 2445 -1.3 13.9 -6.0 -9.4Chile 392 3.7 1429 -0.1 -1.8 -4.6 -4.3Paraguay 206 1.9 384 12.7 3.5 16.8 4.6Uruguay 215 2.4 516 -3.8 1.1 -3.3 0.9* Data for 1993-97 (former Ethiopia).** Data for China include figures for Hong Kong.n.a. not available.

World Wheat Facts and Trends 1998/9935Growth of wheat yield (%/yr) Growth of wheat production (%/yr) Wheat area as percentof total cereal area1951-66 1966-77 1977-85 1985-97 * 1951-66 1966-77 1977-85 1985-97 * (average) 1995-97 (%)0.1 4.1 0.7 2.6 2.2 5.1 1.0 1.5 82.0 3.3 1.5 -2.4 3.0 -2.5 5.5 7.7 121.2 1.1 1.9 0.2 0.3 0.0 1.5 1.6 9-1.9 5.2 0.4 4.0 1.1 8.6 1.3 -0.9 24-0.3 -0.1 5.2 2.0 8.0 13.4 -12.7 14.6 4n.a. 8.0 1.2 -3.0 n.a. 35.6 14.7 9.5 27.1 4.4 4.5 -3.1 10.5 24.3 -2.9 -2.2 20.0 1.2 4.4 2.7 0.0 2.1 3.2 4.0 52-0.9 0.3 4.4 2.1 -0.7 1.8 1.1 0.9 652.3 3.4 1.9 2.9 0.5 3.5 0.8 9.6 391.8 0.9 4.6 -3.3 2.5 0.7 5.0 -1.6 514.4 0.7 5.7 2.2 2.9 3.5 5.4 3.3 680.2 2.8 1.2 1.1 2.8 4.1 1.1 1.8 64-0.1 1.7 0.7 0.8 1.2 2.8 -3.5 2.5 77-1.3 2.2 0.3 4.1 3.8 3.3 1.9 4.4 721.5 0.5 3.7 -1.6 2.2 -0.5 2.1 2.4 481.3 -0.4 9.7 1.5 7.9 -1.8 38.7 -3.6 521.8 2.8 3.8 4.9 1.7 8.1 0.2 8.5 490.4 3.5 0.4 -0.1 3.0 4.8 0.1 0.1 670.5 -0.8 -1.9 1.0 6.5 9.0 -3.6 4.3 141.1 4.3 3.4 2.5 3.2 7.8 5.4 3.5 260.9 6.0 3.4 0.2 4.3 13.9 20.7 1.4 61.4 4.0 3.9 2.7 3.7 8.3 5.5 3.7 254.4 2.6 10.2 -6.4 18.9 -1.3 15.1 -7.4 22.1 -0.7 2.4 2.1 1.1 8.2 6.6 4.6 200.5 5.1 1.8 1.9 2.2 6.2 3.8 2.9 670.9 4.4 8.3 2.5 0.7 5.7 8.6 2.5 320.9 4.4 8.4 2.5 0.6 5.8 8.6 2.6 322.8 5.0 9.2 -6.7 27.5 4.3 12.1 -10.1 99-7.8 8.4 2.1 -1.2 0.9 2.1 2.1 -2.0 67.3 3.8 3.0 0.3 8.3 3.8 9.5 -2.8 67.3 3.9 2.6 0.2 8.4 3.8 9.6 -2.7 80.7 -0.1 2.2 0.1 0.5 -2.8 1.3 1.5 60.1 3.1 1.7 1.6 6.5 5.5 2.4 5.5 190.1 0.6 0.8 -0.3 -0.5 -1.3 -2.6 0.5 121.7 0.8 5.7 2.5 1.9 2.8 6.8 0.6 261.8 3.1 3.4 2.7 2.9 1.8 8.6 3.9 58-0.4 -1.2 7.9 0.3 -1.7 12.7 1.8 -9.1 71.9 -1.2 2.4 3.2 1.9 -3.0 -2.2 -1.1 621.4 -0.7 1.6 2.2 14.1 2.8 18.4 6.8 350.2 0.1 9.3 4.5 -3.6 1.3 6.0 5.4 35

36Selected Wheat StatisticsProduction statistics (cont'd.)Average wheat area, yield, and production, 1995-97Growth of wheat area (%/yr)REGION / COUNTRY Harvested area (000 ha) Yield (t/ha) Production (000 t) 1951-66 1966-77 1977-85 1985-97 *Eastern Europe and former Soviet Union 57159 1.8 100204 2.4 -1.2 -2.5 -0.4Albania 134 2.7 355 1.0 4.0 -0.7 -4.0Azerbaijan 468 1.6 741 n.a. n.a. n.a. 1.8Bulgaria 1117 2.6 2931 -1.7 -2.1 2.7 0.4Croatia 212 3.9 817 n.a. n.a. n.a. 2.9Czech Republic 818 4.6 3730 n.a. n.a. n.a. 1.0Estonia 45 2.1 97 n.a. n.a. n.a. 0.6Hungary 1183 3.3 3928 -2.1 1.0 1.0 -1.7Kazakhstan 12104 0.6 7708 n.a. n.a. n.a. -3.1Kyrgyzstan 438 2.4 1039 n.a. n.a. n.a. 12.7Latvia 137 2.4 332 n.a. n.a. n.a. 3.3Lithuania 328 2.8 900 n.a. n.a. n.a. 3.3Macedonia 121 2.6 315 n.a. n.a. n.a. 0.6Moldova Republic 395 2.9 1159 n.a. n.a. n.a. 7.0Poland 2481 3.4 8479 0.4 0.6 -0.4 2.2Romania 2213 2.7 5989 0.7 -2.4 0.6 -1.0Russian Federation 25224 1.4 36431 n.a. n.a. n.a. 1.6Slovakia 421 4.4 1846 n.a. n.a. n.a. 0.1Slovenia 35 4.1 144 n.a. n.a. n.a. -5.7Turkmenistan 506 1.1 573 n.a. n.a. n.a. 21.6Ukraine 5960 2.7 16075 n.a. n.a. n.a. 1.2Uzbekistan 1317 2.1 2721 n.a. n.a. n.a. 18.1Yugoslavia, Fed. Rep. of 749 3.3 2459 n.a. n.a. n.a. -1.4Western Europe, Japan, and otherhigh-income countries 64474 3.2 204835 -0.1 0.6 1.8 -0.4Australia 10192 1.9 19615 5.1 0.2 2.4 -0.7Austria 254 5.1 1298 2.5 -1.1 1.9 -2.5Belgium-Luxembourg 216 7.9 1700 1.0 -1.2 0.0 0.9Canada 11601 2.3 26337 1.1 -1.0 4.2 -1.9Denmark 660 7.2 4734 4.2 2.3 15.1 6.5Finland 112 3.9 434 5.2 -2.4 4.7 -3.8France 4965 6.8 33559 0.1 0.4 2.7 0.1Germany 2634 7.2 18837 1.5 2.1 0.2 0.8Greece 862 2.4 2063 1.5 -2.0 -1.0 -0.3Ireland 83 8.3 693 -4.3 -4.7 6.2 2.2Italy 2419 3.1 7564 -0.8 -2.9 0.1 -2.7Netherlands 138 8.3 1141 5.1 -2.3 1.0 1.0New Zealand 51 5.5 281 5.6 -2.0 -4.4 -3.9Norway 61 4.5 278 -13.0 20.8 7.7 4.2Spain 2063 2.2 4608 -0.2 -4.5 -2.7 -0.6Sweden 311 6.0 1880 -3.4 4.9 -0.5 0.3Switzerland 100 6.3 635 0.8 -2.0 1.1 0.4United Kingdom 1957 7.7 15143 0.2 2.5 6.9 -0.2United States 25286 2.5 62974 -1.9 3.1 0.7 0.4Regional aggregatesDeveloping countries 104865 2.6 274193 1.1 1.9 0.8 0.4Eastern Europe and former Soviet Union 57159 1.8 100204 2.4 -1.2 -2.5 -0.4Western Europe, Japan, andother high-income countries 64474 3.2 204835 -0.1 0.6 1.8 -0.4World** 226498 2.6 579232 1.2 0.5 0.1 -0.1* Data for 1993-97 (former Czechoslovakia) and 1992-97 (former Soviet Union and former Yugoslavia).** The world aggregates are not exactly equal to the FAO estimates because the method of aggregation may have differed.n.a. not available.

World Wheat Facts and Trends 1998/9937Growth of wheat yield (%/yr) Growth of wheat production (%/yr) Wheat area as percentof total cereal area1951-66 1966-77 1977-85 1985-97 * 1951-66 1966-77 1977-85 1985-97 * (average) 1995-97 (%)1.5 1.8 -0.2 -1.2 3.9 0.6 -2.8 -1.6 48-0.2 6.1 3.8 -1.8 0.8 10.1 3.1 -5.8 58n.a. n.a. n.a. -3.5 n.a. n.a. n.a. -1.6 753.1 3.3 -1.0 -3.8 1.4 1.3 1.7 -3.4 57n.a. n.a. n.a. -0.7 n.a. n.a. n.a. 2.3 34n.a. n.a. n.a. 1.0 n.a. n.a. n.a. 2.0 51n.a. n.a. n.a. 3.1 n.a. n.a. n.a. 3.6 152.7 5.3 3.2 -4.4 0.6 6.4 4.2 -6.0 42n.a. n.a. n.a. -11.6 n.a. n.a. n.a. -14.7 70n.a. n.a. n.a. -0.8 n.a. n.a. n.a. 11.9 71n.a. n.a. n.a. 1.5 n.a. n.a. n.a. 4.7 31n.a. n.a. n.a. 1.9 n.a. n.a. n.a. 5.2 30n.a. n.a. n.a. 0.1 n.a. n.a. n.a. 0.7 53n.a. n.a. n.a. -4.5 n.a. n.a. n.a. 2.5 433.8 3.3 2.5 -0.9 4.2 3.9 2.1 1.2 282.9 4.6 -0.4 -1.1 3.5 2.2 0.2 -2.2 36n.a. n.a. n.a. -4.3 n.a. n.a. n.a. -2.7 48n.a. n.a. n.a. 1.8 n.a. n.a. n.a. 2.0 50n.a. n.a. n.a. -0.1 n.a. n.a. n.a. -5.8 35n.a. n.a. n.a. -16.4 n.a. n.a. n.a. 5.2 78n.a. n.a. n.a. -5.7 n.a. n.a. n.a. -4.5 47n.a. n.a. n.a. 9.8 n.a. n.a. n.a. 27.9 74n.a. n.a. n.a. -0.2 n.a. n.a. n.a. -1.6 322.2 1.6 2.7 1.6 2.1 2.2 4.5 1.2 471.0 0.1 2.0 2.4 6.2 0.2 4.4 1.7 652.3 2.4 3.6 0.7 4.9 1.3 5.4 -1.8 311.3 2.5 4.6 2.2 2.3 1.3 4.6 3.1 690.3 1.7 -0.9 2.3 1.4 0.7 3.3 0.4 601.0 1.2 3.7 1.4 5.2 3.6 18.8 7.9 441.3 3.3 5.5 3.3 6.5 0.9 10.1 -0.5 113.4 2.6 3.8 1.6 3.5 3.0 6.5 1.6 571.4 1.6 3.4 1.9 2.9 3.7 3.6 2.6 392.8 3.1 -0.1 -0.3 4.3 1.1 -1.1 -0.6 662.1 1.4 4.8 2.1 -2.2 -3.3 11.0 4.4 291.8 1.0 2.3 1.4 1.0 -1.9 2.4 -1.2 571.3 2.2 3.4 1.4 6.4 -0.1 4.4 2.5 701.6 -0.1 3.2 3.2 7.1 -2.1 -1.1 -0.6 322.4 2.7 2.7 1.1 -10.6 23.4 10.5 5.3 181.1 2.4 5.3 -1.2 0.9 -2.1 2.6 -1.8 304.3 2.4 3.4 1.4 0.9 7.3 2.9 1.7 261.4 1.3 4.7 1.4 2.2 -0.7 5.8 1.8 512.8 1.6 4.4 1.8 3.0 4.1 11.3 1.7 583.3 1.1 3.0 0.5 1.4 4.2 3.7 0.9 401.0 3.4 5.1 2.2 2.0 5.3 5.9 2.5 231.5 1.8 -0.2 -1.2 3.9 0.6 -2.8 -1.6 482.2 1.6 2.7 1.6 2.1 2.2 4.5 1.2 471.5 2.1 3.0 1.3 2.6 2.6 3.1 1.2 32

38Selected Wheat StatisticsConsumption statisticsAverage net wheat imports, 1995-97 Wheat consumption Average percent wheat useAverage Growth rate HumanTotal Per capita per capita, per capita, consumption Animal feedREGION / COUNTRY (000 t) (kg/yr) 1994-96 (kg/yr) 1987-96 (%/yr) * 1994-96 (%) 1994-96 (%)Eastern and Southern Africa 2619 8 29 1.3 93 1Angola 46 4 23 6.5 99 ++Eritrea 119 36 63 -2.6 95 ++Ethiopia 331 6 39 -1.4 91 ++Kenya 345 12 22 3.0 96 ++Mauritius 110 98 95 2.8 94 ++Mozambique 169 10 13 3.6 99 ++Somalia 10 1 6 -15.7 97 ++South Africa 837 20 69 -0.5 93 2Sudan 216 8 44 2.6 95 ++Tanzania 84 3 5 -2.6 95 ++Zambia 30 4 11 1.2 97 ++Zimbabwe 108 9 28 -1.7 94 ++Western and Central Africa 1918 7 9 0.8 96 1Cameroon 43 3 12 -11.5 97 ++Congo, Dem. Rep. of 142 3 5 -5.1 96 ++Côte d’Ivoire 233 17 17 -3.0 98 ++Ghana 140 8 8 -4.6 98 ++Guinea n.a. n.a. 14 -3.9 98 ++Mauritania 72 31 74 -0.1 97 ++Nigeria 766 7 7 13.5 94 2Senegal 195 23 25 1.7 98 ++North Africa 12847 96 205 0.5 83 5Algeria 2995 104 227 0.9 91 1Egypt 5993 95 180 -0.1 81 8Libya 340 61 262 1.6 63 21Morocco 2280 84 219 1.1 84 3Tunisia 1239 135 234 -0.2 86 1West Asia 8687 37 210 -1.1 78 4Afghanistan 120 6 131 -2.6 92 ++Iran 4330 62 200 0.6 87 6Iraq 918 45 103 -12.1 87 ++Jordan 453 104 151 -0.4 94 ++Lebanon 386 125 167 0.4 74 5Saudi Arabia -198 -11 127 -1.2 95 ++Syria -407 -28 272 1.8 79 7Turkey 1899 31 320 -0.4 63 5Yemen 967 62 126 1.6 98 ++South Asia 4753 4 67 0.9 88 1Bangladesh 1078 9 22 -3.3 94 ++India 366

World Wheat Facts and Trends 1998/9939Consumption statistics (cont'd.)Average net wheat imports, 1995-97 Wheat consumption Average percent wheat useAverage Growth rate HumanTotal Per capita per capita, per capita, consumption Animal feedREGION / COUNTRY (000 t) (kg/yr) 1994-96 (kg/yr) 1987-96 (%/yr) * 1994-96 (%) 1994-96 (%)Southeast Asia and Pacific 7706 17 18 7.5 95 3Indonesia 3927 20 20 10.0 98 ++Malaysia 1092 53 43 1.7 68 23Papua New Guinea 108 24 29 10.7 99 ++Philippines 1801 26 26 4.5 100 ++Thailand 537 9 11 11.3 99 ++Viet Nam 131 2 5 5.2 99 ++East Asia 10950 8 93 0.3 86 4China** 8247 7 94 0.4 86 3Mongolia 0 0 173 -8.5 68 9North Korea 73 3 28 -0.8 92 2South Korea*** 2630 58 74 -1.3 66 33Mexico, Central America, and Caribbean 3610 23 51 -1.5 82 12Costa Rica 185 53 51 -0.1 96 ++Cuba 714 65 83 -7.5 73 20Dominican Republic 262 33 34 0.3 98 ++El Salvador 168 29 39 7.4 99 ++Guatemala 288 26 29 3.1 99 ++Haiti 9 1 32 2.4 95 ++Honduras 138 24 30 3.4 90 10Jamaica 58 23 51 -6.6 98 ++Mexico 1356 15 54 -0.9 78 15Nicaragua 86 20 24 -0.6 98 ++Panama 97 36 43 5.1 97 ++Trinidad and Tobago 141 109 77 -3.7 95 ++Andean Region, South America 3670 36 40 -1.0 97 0Bolivia 181 24 59 1.1 90 ++Colombia 999 27 29 2.8 98 ++Ecuador 383 33 28 -6.7 98 0Peru 1052 44 48 -1.3 97 ++Venezuela 989 44 46 -2.9 97 0Southern Cone, South America 193 1 75 0.1 85 6Argentina -6395 -182 139 -1.1 84 4Brazil 6216 39 54 0.9 90 4Chile 552 38 136 -1.4 87 7Paraguay -93 -19 81 4.2 26 59Uruguay -87 -27 143 3.9 62 26Eastern Europe andformer Soviet Union 1164 3 262 -4.0 51 29Albania 113 33 215 0.7 57 16Armenia 232 64 158 -2.9 91 2Azerbaijan 114 15 177 -4.4 91 1Belarus 634 61 146 2.6 43 37Bosnia Herzegovina 84 23 127 5.5 90 3Bulgaria -55 -7 348 -6.9 43 37Croatia -22 -5 150 -0.3 57 31Czech Republic -312 -30 350 14.8 30 59Estonia 23 15 103 2.6 70 21* Data for 1993-97 (former Czechoslovakia), 1992-97 (former Soviet Union and former Yugoslavia).** Data for China include figures for Hong Kong.*** South Korea is a high-income country but is included here for greater geographical consistency with previous Wheat Facts and Trends.++ Not applicable.

40Selected Wheat StatisticsConsumption statistics (cont'd.)Average net wheat imports, (1995-97) Wheat consumption Average percent wheat usedAverage Growth rate HumanTotal Per capita per capita, per capita, consumption Animal feedREGION / COUNTRY (000 t) (kg/yr) 1994-96 (kg/yr) 1987-96 (%/yr) * 1994-96 (%) 1994-96 (%)Eastern Europe andformer Soviet Union (cont'd.)Georgia 293 54 146 -2.6 96 ++Hungary -1337 -133 283 -7.2 39 42Kazakhstan -2392 -142 464 -3.2 46 15Kyrgyzstan 103 23 250 -6.5 71 14Latvia 57 23 126 3.4 72 10Lithuania -3 -1 230 1.5 56 27Macedonia 93 43 191 -4.1 55 4Moldova Republic -25 -6 228 -12.8 57 23Poland 1219 32 229 -2.0 48 39Romania -781 -34 224 -4.6 62 13Russian Federation 1815 12 273 -6.5 47 32Slovakia -171 -32 345 -3.6 31 59Slovenia 104 54 144 -8.9 53 21Tajikistan 326 55 192 -1.0 98 ++Turkmenistan 205 49 289 -11.7 57 32Ukraine -594 -12 310 -8.2 43 35Uzbekistan 1492 64 182 1.7 96 ++Yugoslavia, Fed. Rep. of -72 -5 234 -7.7 45 17Western Europe, Japan, andother high-income countries -58005 -68 155 1.3 53 35Australia -13921 -771 199 -0.7 37 20Austria -265 -33 131 1.2 49 43Belgium-Luxembourg 2216 210 223 4.0 41 38Canada -17388 -586 269 1.9 31 54Denmark -1038 -198 623 8.0 12 81Finland 157 31 88 -1.6 60 20France -14802 -254 302 4.1 32 57Germany -2778 -34 176 0.4 36 53Greece 366 35 168 -1.1 82 4Ireland 225 63 265 2.4 39 52Israel 829 147 221 3.3 60 22Italy 5975 104 183 -0.6 80 13Japan 6069 48 53 0.6 87 7Netherlands 2238 144 156 4.6 36 50New Zealand 207 57 119 1.8 62 18Norway 188 43 119 0.3 78 19Portugal 1073 109 129 1.6 75 14Spain 2284 58 140 -0.2 65 22Sweden -287 -33 178 5.6 40 48Switzerland 242 33 130 0.6 70 24United Kingdom -2446 -42 201 -0.1 43 45United States -28096 -104 125 1.2 70 22Regional aggregatesDeveloping countries 56953 13 74 0.3 85 4Eastern Europe andformer Soviet Union 1164 3 262 -4.0 51 29Western Europe, Japan, andother high-income countries -58005 -68 155 1.3 53 35World** — — 100 -0.7 71 16* Data for 1993-97 (former Czechoslovakia), 1992-97 (former Soviet Union and former Yugoslavia).** The world aggregates are not exactly equal to the FAO estimates because the method of aggregation may have differed.++ Not applicable.

World Wheat Facts and Trends 1998/9941CIMMYT wheat experimental andnational average yield, 1995-97 (t/ha)Experimental National averageCOUNTRY/REGION * wheat yield wheat yieldExperimental National averageCOUNTRY/REGION * wheat yield wheat yieldKenya 2.9 1, 5, 7 1.9Madagascar 6.8 1 2.0Malawi 1.5 0.7South Africa 7.3 1.7Sudan 3.0 2, 3 1.9Swaziland 9.9 4 1.5Tanzania 2.2 1.9Zimbabwe 6.6 4.1Eastern and Southern Africa 5.0 2.0Algeria 2.5 1.1Egypt 6.0 5.6Lybia 9.4 1 1.2Morocco 4.1 1.2Tunisia 3.1 1.4North Africa 5.0 2.1Afghanistan 4.0 1.3Iran 5.0 1.6Jordan 2.3 1 1.4Saudi Arabia 6.4 4.8Syria 3.7 2, 3 2.4Turkey 8.7 2.0Yemen Democratic Rep. 1.5 6 1.2West Asia 4.5 2.1Bangladesh 4.4 2.0India 4.4 2.6Myanmar 2.4 0.9Nepal 1.8 1.6Pakistan 3.3 2.1South Asia 3.3 1.8Thailand 2.2 0.7Vietnam 2.3Southeast Asia and the Pacific 2.2 0.7China 6.0 3.8South Korea 3.6 7 4.0Taiwan 1.4 1East Asia 4.8 3.9* Regional aggregates include only countries that have been reported.Guatemala 5.3 2.1Mexico 5.3 1, 5 4.1Mexico,Central America,and the Caribbean 5.3 3.1Bolivia 2.9 0.9Colombia 2.4 1, 3 1.9Ecuador 2.5 0.7Peru 5.4 1.2Andean Region 3.3 1.2Argentina 3.0 2.3Brazil 3.6 1.7Chile 7.2 3.6Paraguay 3.3 1.9Uruguay 3.2 2.4South America 4.1 2.4Bulgaria 5.1 3 2.9Czech Rep. 7.3 4.6Poland 5.3 3.4Romania 2.4 5 1.8Ukraine 3.5 2.7Yugoslavia 3.4 1 4.3Eastern Europe andformer Soviet Union 4.5 3.3Canada 4.0 2.3France 7.1 6.8Greece 3.4 1, 4 2.3Italy 5.6 3.1Japan 3.5 5 3.0New Zealand 6.1 5.5Portugal 2.5 1.3Qatar 6.2 2 2.3Spain 4.6 2.2United Kingdom 7.4 2 6.8Western Europe, North America,and other high-income countries 4.8 3.6Total 4.4 2.5Notes: 1 1990, 2 1992, 3 1993, 4 1995, 5 1996, 6 1997, 7 1998

42Selected Wheat StatisticsWheat area by type of wheat (%)Wheat area by type of wheat, 1997 (%)Wheat area undersemidwarf wheatCOUNTRY/REGION* Spring bread Spring durum Winter bread Winter durum varieties, 1997 (%)Ethiopia 60 40 0 0 51Kenya 100 0 0 0 100South Africa 55 0 45 0 70Sudan 100 0 0 0 80Tanzania 100 0 0 0 98Zambia 100 0 0 0 100Zimbabwe 100 0 0 0 100Eastern and Southern Africa 65 16 19 0 66Nigeria 100 0 0 0 99Western and Central Africa 100 0 0 0 99Algeria 44 56 0 0 71Egypt 94 6 0 0 98Morocco 57 43 0 0 95Tunisia 15 85 0 0 89North Africa 54 46 0 0 88Afghanistan 0 0 100 0 31Iran 48 0 52 0 59Jordan 21 79 0 0 57Lebanon 0 100 0 0 83Syria 39 61 0 0 95Turkey 19 18 50 13 58Yemen 100 0 0 0 43West Asia 30 14 51 6 59Bangladesh 100 0 0 0 100India 100 0 0 0 92Nepal 100 0 0 0 92Pakistan 100 0 0 0 94South Asia 100 0 0 0 92China 44 0 56 0 79East Asia 44 0 56 0 79Guatemala 100 0 0 0 100Mexico 53 47 0 0 99Mexico, Central America, and the Caribbean 54 46 0 0 99Bolivia 90 10 0 0 80Colombia 100 0 0 0 99Ecuador 100 0 0 0 100Peru 83 17 0 0 83Andean Region 90 10 0 0 85Argentina 100 0 0 0 98Brazil 100 0 0 0 53Chile 41 14 45 0 95Paraguay 100 0 0 0 100Uruguay 76 0 24 0 85South America 97 1 3 0 89TOTAL 66 6 26 1 81* Regional aggregates include only countries that have been reported.

World Wheat Facts and Trends 1998/9943Prices and input use for wheatConsumer Fertilizer Farm wageFarm prices of price of Ratio of farm-level fertilizer Fertilized applied per in kg ofwheat, 1998-99 wheat flour, price to wheat price 1998-99 area, hectare of wheat wheatBread Durum 1998-99 1998 harvested, 1998 per day,REGION / COUNTRY (US$/t) (US$/t) Nitrogen Phosphorus Potassium (%) (kg nutrients/ha) 1998Eastern and Southern AfricaEthiopia 173 173 443 2.2 ++ ++ 84 67 5Sudan 194 ++ 400 3.6 ++ ++ 100 190 4Tanzania 209 ++ 468 2.6 ++ ++ 1 n.a. 10Zimbabwe 316 ++ 516 n.a. ++ ++ 100 300 3North AfricaAlgeria 262 292 462 3.9 2.9 ++ 8 500 57West AsiaSaudi Arabia 400 400 545 4.0 2.0 7.5 n.a. n.a. n.a.Turkey 159 208 524 3.3 4.2 7.4 98 198 84South AsiaBangladesh 179 ++ 333 1.8 100 125 6India 131 148 180 4.4 3.4 2.1 94 156 10Nepal 139 ++ 241 6.8 4.6 ++ 80 100 6East AsiaChina 133 224 234 3.5 2.7 2.1 97 246 8Mongolia 78 88 299 3.3 3.0 2.3 30 35 19Mexico, Central America,and the CaribbeanMexico 137 147 263 2.7 3.5 ++ 80 240 38Andean RegionBolivia 158 ++ 344 11.1 6.6 4.9 11 71 31Colombia 226 ++ 566 8.7 4.8 17.8 30 200 34Ecuador 317 ++ 514 2.5 2.9 1.8 40 160 27South AmericaArgentina 115 ++ 390 13.0 7.5 90 66 108Brazil 94 191 244 31.3 10.8 14.7 92 162 73Chile 194 194 536 2.8 3.2 2.4 85 140 48Uruguay 105 ++ 636 9.3 5.6 ++ 100 100 155Eastern Europe andformer Soviet UnionBulgaria 70 80 140 5.2 ++ ++ 70 100 78Western Europe, North America,and other high-income countriesCanada 115 161 665 5.9 5.0 ++ 70 50 499Spain 152 163 227 5.1 5.5 6.0 95 400 245Finland 136 ++ 399 6.0 16.4 ++ 100 600 55France 112 ++ 573 15.2 9.7 9.7 100 380 629Germany 115 184 686 4.0 10.4 6.0 100 315 668Greece 166 163 1000 4.3 12.2 ++ 95 14 184Ireland 70 ++ 365 4.2 ++ ++ 100 265 401Italy 161 158 421 5.9 ++ ++ 94 290 519Netherlands 135 ++ 1779 5.4 9.0 7.1 100 180 692Portugal 147 164 352 11.6 6.6 8.5 100 230 180Switzerland 473 ++ 1190 1.5 ++ ++ 95 324 115United Kingdom 148 ++ 1066 22.9 ++ ++ 98 190 444United States 106 117 1326 7.2 ++ ++ 95 140 528++ not applicable.n.a. not available.

44ReferencesReferencesAbdElGhani, A.M., A.M. AbdElShafi, and E.H.Ghanem. 1994. Genetic variability anddifferential plant adaptations as indicators ofwheat potential under heat stress in upperEgypt. In D.A. Saunders and G.P. Hettel (eds.),Wheat in Heat Stressed Environments:Irrigated, Dry Areas and Rice-Wheat FarmingSystems. Mexico, D.F.: CIMMYT.Alemu Hailye, H. Verkuijl, W. Mwangi, andAsmare Yellew. 1998. Farmers’ Wheat SeedSources and Seed Management in EnebssieArea, Ethiopia. Mexico, D.F.: IAR and CIMMYT.Bohn, A., and D. Byerlee. 1993. The wheatbreeding industry in developing countries: Ananalysis of investments and impacts. Part I of1992/93 CIMMYT World Wheat Facts andTrends. The Wheat Breeding Industry inDeveloping Countries: An Analysis ofInvestments and Impacts. Singapore: CIMMYT.Bohn, A., D. Byerlee, and M.K. Maredia.Forthcoming. Investment in wheatimprovement in developing countries. In M.K.Maredia and D. Byerlee (eds.), The GlobalWheat Improvement System: Prospects forEnhancing Efficiency in the Presence ofSpillovers. CIMMYT Research Report No. 5.Mexico, D.F.: CIMMYT.Borlaug, N.E. 1968. Wheat breeding and itsimpact on world food supply. In K.W. Finlayand K.W. Shephard (eds.), Proceedings of the3rd International Wheat Genetics Symposium.Canberra, Australia: Australian Academy ofSciences. Pp. 1-36._______. 1972. A cereal breeder and exforester’sevaluation of the progress andproblems involved in breeding rust resistantforest trees: Moderator’s Summary. In Biologyof Rust Resistance in Forest Trees: Proceedingsof a NATO-IUFRO Advanced Study Institute,August 17-24, 1969. Washington, D.C.: USDAForest Service Misc. Publication 1221. Pp. 615-642.Bramel-Cox, P.J., T. Barker, F. Zavala-García, andJ.D. Eastin. 1991. Selection and testingenvironments for improved performanceunder reduced-input conditions. In PlantBreeding and Sustainable Agriculture:Considerations for Objectives and Methods.CSSA Special Publication No. 18. Madison,Wisconsin: Crop Science Society of Americaand American Society of Agronomy. Pp. 29-56.Brennan, J.P., and D. Byerlee. 1991. The rate ofcrop varietal replacement on farms: measuresand empirical results for wheat. Plant Varietiesand Seeds 4: 99-106.Byerlee, D., and M. Morris. 1993. Research formarginal environments: Are weunderinvested? Food Policy 18(3): 381-393.Byerlee, D., and P. Moya. 1993. Impacts ofInternational Wheat Breeding Research in theDeveloping World, 1969-1990. Mexico, D.F.:CIMMYT.Byerlee, D., and G. Traxler. 1995. National andinternational wheat improvement research inthe post green revolution period: Evolutionand impacts. American Journal of AgriculturalEconomics 77: 268-278._______. 1996. The role of technology spilloversand economies of size in the efficient designof agricultural research systems. In ConferenceProceedings, Global Agricultural Science Policyfor the Twenty-First Century, Invited Papers,26-28 August, 1996. Melbourne: TheConference Secretariat.Caldwell, R.M. 1968. Breeding for general and/orspecific plant disease resistance. In K.W. Finlayand K.W. Shephard (eds.), Proceedings of the3rd International Wheat Genetics Symposium.Canberra, Australia: Australian Academy ofSciences. Pp. 263-272.Calhoun, D.S., G. Gebeyehu, A. Miranda, S.Rajaram, and M. van Ginkel. 1994. Choosingevaluation environments to increase wheatgrain yield under drought conditions. CropScience 34: 673-678.CIMMYT. 1991. 1990-91 CIMMYT World WheatFacts and Trends: Wheat and BarleyProduction in Rainfed Marginal Environmentsof the Developing World. Mexico, D.F.:CIMMYT._______. 1992. CIMMYT 1991 Annual Report.Improving the productivity of maize and wheatin developing countries: An assessment ofimpact. Mexico, D.F.: CIMMYT._______. 1993. 1992-93 World Wheat Facts andTrends. The Wheat Breeding Industry inDeveloping Countries: An Analysis ofInvestment and Impacts. Mexico, D.F.:CIMMYT._______.1996. World Wheat Facts and Trends1995/96: Understanding Global Trends in theUse of Wheat Diversity and International Flowsof Wheat Genetic Resources. Mexico, D.F.:CIMMYT._______. 1997a. CIMMYT in 1996-97. Wideningthe Circle of Partnerships. Mexico,D.F.:CIMMYT._______. 1997b. Book of Abstracts. The Geneticsand Exploitation of Heterosis in Crops; AnInternational Symposium. Mexico, D.F.:CIMMYT.Cooper, M., D.E. Byth, and D.R. Woodruff. 1994.An investigation of the grain yield adaptationof advanced CIMMYT wheat lines to waterstress environments in Queensland. II.Classification analysis. Australian Journal ofAgriculture 45: 985-1002.Çukadar, B, R.J. Peña, D.Dunphy, and M. vanGinkel. 1997. The potential of hybrid wheatunder irrigated conditions in Mexico. In Bookof Abstracts. The Genetics and Exploitation ofHeterosis in Crops. An InternationalSymposium. Mexico, D.F.: CIMMYT. Pp. 190-191.David, C., and K. Otsuka. 1992. The GreenRevolution and income distribution acrossproduction environments in Asia. Paperpresented at the Rice Research Seminar,International Rice Research Institute (IRRI), LosBaños, Philippines.Dubin, H.J., and S. Rajaram. 1996. Breedingdisease-resistant wheats for tropical highlandsand lowlands. Annual Review ofPhytopathology 34: 503-526.Duvick, D.N. 1990. Ideotype evolution of hybridmaize in the USA, 1930-1990. In Vol. II of ATTIProceedings, National Maize Conference:Research, Economy, Environment. Bologna,Italy: Centro Regionale per la SperimentazioneAgraria, Pozzuolo del Friuli Edagrocole s.p.a.Pp. 557-570.Duvick, D.N. 1992. Genetic contributions toadvances in yield of US maize. Maydica 37: 69-79.Ehdaie, B., J.G. Waines, and A.E. Hall. 1988.Differential responses of landrace andimproved spring wheat genotypes to stressenvironments. Crop Science 28: 838-842.Evans, L.T., and R.A. Fischer. Forthcoming. Yieldpotential: Its definition, measurement andsignificance. In Symposium on Post-GreenRevolution Trends in Crop Yield Potential:Increasing, Stagnant, or Greater Resistance toStress. To be published in Crop Science.FAO (Food and Agriculture Organization of theUnited Nations). 1999. FAOSTAT Food BalanceSheets. Rome, Italy: FAO._______. 1999. FAOSTAT Production Statistics.Rome, Italy: FAO._______. 1999. FAOSTAT Trade Statistics. Rome,Italy: FAO.Hailu Beyene, H. Verkuijl, and W. Mwangi. 1998.Farmers’ Seed Sources and Management ofBread Wheat in Wolmera Woreda, Ethiopia.Mexico, D.F.: CIMMYT and IAR.Heisey, P.W. (ed.). 1990. Accelerating the Transferof Wheat Breeding Gains to Farmers: A Studyof the Dynamics of Varietal Replacement inPakistan. CIMMYT Research Report No. 1.Mexico, D.F.:CIMMYTHeisey, P.W., and J.P. Brennan. 1991. An analyticalmodel of farmers’ demand for replacementseed. American Journal of AgriculturalEconomics 73: 1044-1052.Heisey, P., and M. Lantican. Forthcoming (1999).International wheat breeding research inEastern and Southern Africa, 1966–1997. InThe Tenth Regional Wheat Workshop forEastern, Central, and Southern Africa. AddisAbaba, Ethiopia: CIMMYT.Hoogendoorn, J., W.H. Pfeiffer, S. Rajaram, andM.D. Gale. 1988. Adaptive aspects of dwarfinggenes in CIMMYT germplasm. In T.E. Miller andR.M.D. Koebner (eds.), Proceedings of the 7 thInternational Wheat Genetics Symposium.Cambridge, UK. Pp.1093-1100.ICARDA (International Center for AgriculturalResearch in the Dry Areas). 1993. CerealProgram. Annual Report for 1992. Aleppo,Syria: ICARDA.Jordaan, J.P. 1996. Hybrid wheat: Advances andchallenges. In M.P. Reynolds, S. Rajaram, andA. McNab (eds.), Increasing Yield Potential inWheat in Wheat: Breaking the Barriers.Mexico, D.F.: CIMMYT. Pp. 66-75.Kilpatrick, R.A. 1975. New Wheat Cultivars andLongevity of Rust Resistance, 1971–75. PaperARS-NE-4, USDA Agricultural Research Service.Beltsville, Maryland: USDA.Kohli, M.M., and S. Rajaram. 1988. WheatBreeding for Acid Soils: Review of Brazilian/CIMMYT Collaboration, 1974-1986. Mexico,D.F.: CIMMYT.

World Wheat Facts and Trends 1998/9945Longmire, J., and A. Moldashev. 1999. ChangingCompetitiveness of the Wheat Sector ofKazakhstan and Sources of Future ProductivityGrowth. CIMMYT Economics Program WorkingPaper No. 96-06. Mexico, D.F.: CIMMYT.Mann, C.C. 1999. Crop scientists seek a newrevolution. Science 283: 310-314.Maredia, M.K. 1993. The Economics ofInternational Agricultural Research Spillovers:Implications for the Efficient Design of WheatImprovement Research Programs. Ph.D.dissertation, Michigan State University,Department of Agricultural Economics, EastLansing.Maredia, M.K., and D. Byerlee (eds.) 1999. TheGlobal Wheat Improvement System: Prospectsfor Enhancing Efficiency in the Presence ofSpillovers. CIMMYT Research Report No. 5.Mexico, D.F.: CIMMYT.Maredia, M.K., and C.K. Eicher. 1995.TheEconomics of Wheat Research in DevelopingCountries: The One Hundred Million DollarPuzzle. World Development. 23(3):401-412.McIntosh, R.A. 1998. Breeding wheat forresistance to biotic stresses. In H.J. Braun, F.Altay, W.E. Kronstad, S.P.S. Beniwal, and A.McNab (eds.), Wheat: Prospects for GlobalImprovement. Proceedings of the 5 thInternational Wheat Conference. Dordrecht,the Netherlands: Kluwer Academic Publishers.Pp. 71-86.McIntosh, R.A., G.E. Hart, and M.D. Gale. 1995.Catalogue of gene symbols for wheat. In Z.S. Liand Z.Y. Xin (eds.), Proceedings of the 8thInternational Wheat Genetics Symposium.Beijing, China. Pp. 1333-1500.Morgounov, A. 1995. Bread wheat breeding forheat tolerance. In S. Rajaram and G.P. Hettel(eds.), Wheat Breeding at CIMMYT:Commemorating 50 Years of Research inMexico for Global Wheat Improvement. WheatSpecial Report No. 29. Mexico, D.F.: CIMMYT.Pp. 29-35.Morris, M.L., A. Belaid, and D. Byerlee. 1991.Wheat and barley production in rainfedmarginal environments of the developingworld. Part 1 of 1990-91 CIMMYT WorldWheat Facts and Trends. Mexico, D.F.:CIMMYT.Morris, M.L., H.J. Dubin, and T. Pokhrel. 1994.Returns to wheat breeding research in Nepal.Agricultural Economics 10(3): 269 282.Mujeeb-Kazi, A., and G.P. Hettel (eds.). 1995.Utilizing Wild Grass Biodiversity in WheatImprovement: 15 Years of Wide Cross Researchat CIMMYT. CIMMYT Research Report No. 2.Mexico, D.F.: CIMMYT.Ortiz-Monasterio, J.I., K.D. Sayre, S. Rajaram, andM. McMahon. 1997. Genetic progress inwheat yield and nitrogen use efficiency underfour nitrogen rates. Crop Science 37: 898-904.Pardey, P.G., J.M. Alston, J.E. Christian, and S.Fan. 1996. Hidden Harvest: U.S. Benefits fromInternational Research Aid. Food Policy Report.Washington, D.C.: International Food PolicyResearch Institute (IFPRI).Parlevliet, J.E. 1975. Partial resistance of barley toleaf rust, Puccinia hordei. I. Effect of cultivarand development stage on latent period.Euphytica 24: 21-27.Peterson, C.J., J.M. Moffatt, and J.R. Erickson.1997. Yield stability of hybrid versus purelinehard winter wheats in regional performancetrials. Crop Sci. 37:116-120.Picket, A.A. 1993. Hybrid wheat: Result andproblems. Advances in Plant Breeding No. 15.Berlin, Germany: Paul Parey ScientificPublishers.Pingali, P., and P.W. Heisey. 1996. Cereal cropproductivity in developing countries: Pasttrends and future prospects. In GlobalAgricultural Science Policy for the Twenty-FirstCentury. Melbourne, Australia: NaturalResources and Environment. Pp. 51-94.Pingali, P.L., M. Hossain, and R.V. Gerpacio. 1997.Asian Rice Bowls: The Returning Crisis?Wallingford, UK: IRRI and CAB International.Rajaram, S., R.P. Singh, and M. van Ginkel. 1996.Approaches to breeding wheat for wideadaptation, rust resistance, and droughttolerance. Paper presented at the 8 th Assemblyof the Wheat Breeding Society of Australia, 29September-4 October, Canberra, Australia._______. 1997. Breeding wheat for wideadaptation, rust resistance, and droughttolerance. Chapter 8 in M.S. Kang (ed.), CropImrovement for the 21st Century. India:Research Signpost.Rajaram, S., and M. van Ginkel. 1996. Yieldpotential debate: Germplasm vs. methodology,or both. In M.P. Reynolds, S. Rajaram, and A.McNab (eds.), Increasing Yield Potential inWheat in Wheat: Breaking the Barriers.Mexico, D.F.: CIMMYT. pp. 11-18.Rasmusson, D.F. 1996. Germplasm is paramount.In M.P. Reynolds, S. Rajaram, and A. McNab(eds.), Increasing Yield Potential in Wheat inWheat: Breaking the Barriers. Mexico, D.F.:CIMMYT. Pp.28-35.Regassa Ensermu, W. Mwangi, H. Verkuijl, M.Hassena, and Z. Alemayehu. 1998. Farmers’Wheat Seed Sources and Seed Management inChilalo Awraja, Ethiopia. Mexico, D.F.: IAR andCIMMYT.Rejesus, R.M., P.W. Heisey, and M. Smale. 1999.Sources of Productivity Growth in Wheat: AReview of Recent Performance and MediumtoLong-Term Prospects. CIMMYT EconomicsWorking Paper 99-05. Mexico D.F.: CIMMYTRenkow, M. 1991. Modeling the AggregateEffects of technological Change on IncomeDistribution in Pakistan’s Favored and MarginalProduction Environments. CIMMYT EconomicsPaper No.4. Mexico, D.F.: CIMMYT.Reynolds, M.P., M. Balota, M.I.B. Delgado, I.Amani, and R.A. Fischer. 1994. Physiologicaland morphological traits associated with springwheat yield under hot, irrigated conditions.Australian Journal of Plant Physiology 21: 717-30.Rosegrant, M.W., and P.L. Pingali. 1994. Policyand technology for rice productivity growth inAsia. Journal of International Development6(6): 665-688.Rosegrant, M.W., M.A. Sombilla, R.V. Gerpacio,and C. Ringler. 1997. Global food markets andUS exports in the twenty-first century. Paperpresented at the Illinois World Food andSustainable Agriculture Program Conference,Meeting the Demand for Food in the 21 stCentury: Challenges and Opportunities, 28May, University of Illinois, Urbana-Champaign.Sayre, K.D. 1996. The role of crop managementresearch at CIMMYT in addressing breadwheat yield potential issues. In M.P. Reynolds,S. Rajaram, and A. McNab (eds.), IncreasingYield Potential in Wheat in Wheat: Breakingthe Barriers. Mexico, D.F.: CIMMYT. Pp.203-207._______. 1998. Ensuring the Use of SustainableCrop Management Strategies by Small WheatFarmers in the 21 st Century. Wheat SpecialReport No. 48. Mexico, D.F: CIMMYT.Sayre, K.D., S. Rajaram, and R.A. Fischer. 1997.Yield potential progress in short bread wheatsin Northwest Mexico. Crop Science 37: 36-42.Sayre, K.D., R.P. Singh, J. Huerta-Espino, and S.Rajaram. 1998. Genetic progress in reducinglosses to leaf rust in CIMMYT-derived Mexicanspring wheat cultivars. Crop Science 38: 654-659.Sidhu, D.S., and D. Byerlee. 1992. TechnicalChange and Wheat Productivity in the IndianPunjab in the Post-Green Revolution Period.CIMMYT Economics Working Paper 92-02.Mexico, D.F.: CIMMYT.Singh, A.J., and D.Byerlee 1990. Relativevariability in wheat yields across countries overtime. Journal of Agricultural Economics 41:21-32.Snape, J.W. 1998. Golden calves or whiteelephants? Biotechnology for wheatimprovement. Euphytica 100: 207-217.Sorrells, M.E. 1996. Application of molecularmarkers to wheat improvement. In M.P.Reynolds, S. Rajaram, and A. McNab (eds.),Increasing Yield Potential in Wheat: Breakingthe Barriers. Mexico, D.F.: CIMMYT. Pp.182-194.Traxler, G., J. Falck-Zepeda, I. Ortiz-Monasterio,K.D.Sayre.1995. Production risk and evolutionof varietal technology. American Journal ofAgricultural Economics. 77(1):1-7.Traxler G., and P.L. Pingali. 1998. Enhancing thediversity of modern germplasm through theinternational coordination of research roles. InM. Smale (ed.), Farmers, Gene Banks and CropBreeding: Economic Analyses of Diversity inWheat, Maize, and Rice. Boston: KluwerAcademic Publishers and CIMMYT.Uddin, N., B.F. Carver, and A.C. Clutter. 1992.Genetic analysis and selection for wheat yieldin drought-stressed and irrigatedenvironments. Euphytica 62: 89-96.van Ginkel, M., D.S. Calhoun, G. Gebeyehu, A.Miranda, C. Tian-you, R. Pargas Lara, R.M.Trethowan, K. Sayre, J. Crossa, and S. Rajaram.1998. Plant traits related to yield of wheat inearly, late, or continuous drought conditions.Euphytica 100: 109-21.Virmani, S.1999. Exploitation of heterosis forshifting the yield frontier in rice. In J.G. Coors,and S. Pandey (Eds.). Genetics and Exploitationof Heterosis in Crops. USA: American Societyof Agronomy, Inc. and Crop Science Society ofAmerica, Inc.World Bank. 1999. Revival of the Seed Sector inKazakhstan. Washington, D.C.: World Bank.Zavala-Garcia, F., P.J. Bramel-Cox, J.D. Eastin,M.D. Witt, and D.J. Andrews. 1992. Increasingthe efficiency of crop selection forunpredictable environments. Crop Science 32:51-57.

46References Appendix AAppendix ARegions of the WorldDeveloping CountriesEASTERN AND SOUTHERN AFRICASOUTHEAST ASIA AND THE PACIFICAngolaMozambique American Samoa PhilippinesBotswanaNamibiaCook Islands SamoaBurundiRwandaEast TimorSolomon IslandsComorosSeychellesFijiThailandDjiboutiSomaliaIndonesiaTokelauEritreaSouth Africa KiribatiTongaEthiopiaSudanLaosTuvaluKenyaSwazilandMalaysiaVanuatuLesothoTanzaniaNauruVietnamMadagascar UgandaNiue IslandWallis and FutunaMalawiZambiaNorfolk IslandIslandMauritiusZimbabwePapua New GuineaWESTERN AND CENTRAL AFRICAEAST ASIABeninGuineaChinaNorth KoreaBurkina Faso Guinea-Bissau MongoliaSouth KoreaCameroonLiberiaCape VerdeMaliMEXICO, CENTRAL AMERICA,Central Africa MauritaniaAND THE CARIBBEANRepublicNigerAntigua and Barbuda MontserratChadNigeriaBarbadosNetherlandsCongo, Democratic Sao Tome and BelizeAntillesRepublic ofPrincipeCosta RicaNicaraguaCongo, Republic of SenegalCubaPanamaCôte d’Ivoire Sierra Leone DominicaSaint Kitts andEquatorial Guinea Saint Helena Dominican Republic NevisGambiaTogoEl SalvadorSaint LuciaGhanaGrenadaSaint PierreGuadeloupeMiquelonNORTH AFRICAGuatemalaSaint VincentAlgeriaMoroccoHaitiGrenadinesEgyptTunisiaHondurasTrinidad andLibyaJamaicaTobagoWEST ASIAMexicoAfghanistan OmanANDEAN REGION, SOUTH AMERICABahrainSaudi Arabia BoliviaPeruIranSyriaColombiaSurinameIraqTurkeyEcuadorVenezuelaJordanYemenGuyanaLebanonSOUTHERN CONE, SOUTH AMERICASOUTH ASIAArgentinaFalkland IslandsBangladeshMyanmarBrazilParaguayBhutanNepalChileUruguayIndiaPakistanMaldivesSri LankaEastern Europe andFormer Soviet UnionAlbaniaArmeniaAzerbaijanBelarusBosnia HerzegovinaBulgariaCroatiaCzech RepublicEstoniaGeorgiaHungaryKazakhstanKyrgyzstanLatviaLithuaniaMacedoniaMoldova RepublicPolandRomaniaRussianFederationSlovakiaSloveniaTajikistanTurkmenistanUkraineUzbekistanYugoslavia,Fed. Rep. ofWestern Europe, Japan,and Other High-IncomeCountriesAustraliaAustriaBelgium-LuxembourgBrunei DarussalamCanadaCyprusDenmarkFaeroe IslandFinlandFranceGermanyGreeceGreenlandIcelandIrelandIsraelItalyJapanKuwaitMaltaNetherlandsNew ZealandNorwayPortugalQatarSingaporeSpainSwedenSwitzerlandUnited ArabEmiratesUnited KingdomUnited States

ISSN: 0257-876XLisboa 27, Apdo. Postal 6-641, 06600 Mexico, D.F., MexicoCIMMYT home page on

More magazines by this user
Similar magazines