New Wheatsfor a Secure,

Sustainable Future

Timothy G. Reeves, Sanjaya Rajaram,

Maarten van Ginkel, Richard Trethowan,

Hans-Joachim Braun, and Kelly Cassaday

ISBN: 970-648-040-4

International Maize and Wheat Improvement Center Lisboa 27, Apdo. Postal 6-641, 06600 Mexico, D.F., Mexico

CIMMYT home page on www.cimmyt.cgiar.org

Maarten van Ginkel, head of bread wheat

breeding at CIMMYT, holds one of the

large-spiked wheats (right) that promise

to raise yields in wheats. On the left he

holds a normal wheat spike. (See page 7.)

29

New Wheats for

a Secure, Sustainable

Future

Timothy G. Reeves, Sanjaya Rajaram, Maarten van Ginkel,

Richard Trethowan, Hans-Joachim Braun, and Kelly Cassaday*

* All authors are staff of CIMMYT. T.G. Reeves is Director General; S. Rajaram is Directorof the Wheat Program; M. van Ginkel is Head, Bread Wheat Breeding; RichardTrethowan is a Wheat Breeder; H.-J. Braun is a Wheat Breeder (based in Turkey); andKelly Cassaday is a Writer/Editor. An earlier version of this paper was presented at the9th Wheat Breeding Assembly, 27 September-1 October, 1999, University of SouthernQueensland, Toowoomba, Queensland, Australia.

CIMMYT (www.cimmyt.mx or www.cimmyt.cgiar.org) is an internationally funded, nonprofitscientific research and training organization. Headquartered in Mexico, the Center works withagricultural research institutions worldwide to improve the productivity, profitability, andsustainability of maize and wheat systems for poor farmers in developing countries. It is one of 16similar centers supported by the Consultative Group on International Agricultural Research(CGIAR). The CGIAR comprises about 60 partner countries, international and regionalorganizations, and private foundations. It is co-sponsored by the Food and AgricultureOrganization (FAO) of the United Nations, the International Bank for Reconstruction andDevelopment (World Bank), the United Nations Development Programme (UNDP), and the UnitedNations Environment Programme (UNEP). Financial support for CIMMYT’s research agenda alsocomes from many other sources, including foundations, development banks, and public and privateagencies.

CIMMYT supports Future Harvest, a public awareness campaign that buildsunderstanding about the importance of agricultural issues and international

agricultural research. Future Harvest links respected research institutions, influential public figures,and leading agricultural scientists to underscore the wider social benefits of improved agriculture—peace, prosperity, environmental renewal, health, and the alleviation of human suffering(www.futureharvest.org).

International Maize and Wheat Improvement Center (CIMMYT) 1999. Responsibility for thispublication rests solely with CIMMYT. The designations employed in the presentation of materialin this publication do not imply the expressions of any opinion whatsoever on the part of CIMMYTor contributory organizations concerning the legal status of any country, territory, city, or area, or ofits authorities, or concerning the delimitation of its frontiers or boundaries.

Printed in Mexico.

Correct citation: Reeves, T.G., S. Rajaram, M. van Ginkel, R. Trethowan, H-J. Braun, and K.Cassaday. 1999. New Wheats for a Secure, Sustainable Future. Mexico, D.F.: CIMMYT.

Abstract: This paper reviews strategies used by CIMMYT and its partners to develop sustainablewheat production systems for favored and marginal areas. These strategies aim to achieve anoptimal combination of the best genotypes (G), in the right environments (E), under appropriatecrop management (M), and appropriate to the needs of the people (P) who must implement andmanage them. The first section of the paper presents new options for raising wheat yield potentialand discusses research on disease and stress tolerance, which is aimed at protecting yield potentialin farmers’ fields (with special emphasis on drought tolerance). Next, advances in durum wheatyield potential are reviewed; these advances may prove particularly valuable in marginalenvironments. Other wheat research initiatives for marginal environments are described as well.This is followed by a review of the role of biotechnology in wheat improvement, research on wheatquality, and initiatives in crop and natural resource management research. The paper concludeswith a summary of the latest data on the global impacts of wheat research and a discussion oftrends that could affect whether and how this impact is maintained.

ISBN: 970-648-040-4

AGROVOC descriptors: Wheats; Triticum; Hard wheat; Winter crops; Spring crops; High yieldingvarieties; Hybrids; Plant production; Production policies; Food production; Food security; Plantbreeding; Plant biotechnology; Nutrient improvement; Drought resistance; Pest resistance; Diseaseresistance; Crop management; Resource management; Sustainability; Innovation adoption; Yieldincreases; On farm research

Additional keywords: CIMMYT; Participatory research

AGRIS category codes: E14 Development Economics and Policies; F30 Plant Genetics and Breeding

Dewey decimal classificataion: 338.162ii

Contents

iv Acknowledgments

1 New Wheats for a Secure, Sustainable Future

2 Agriculture: An Agent of Change

3 Prerequisites for Sustainable Agriculture

4 Breeding Wheats for Lasting Food Security

4 Options for Increasing Yield Potential

6 Gene pools of winter and spring hexaploid wheats

6 Introgressing spring and winter wheat gene pools

6 Chinese wheats: A wellspring of diversity

6 Hybrid wheats

7 Landraces

7 Improved plant ideotype

7 Phenological traits

7 Physiological traits

8 Synthetic wheats: Delivering diversity to plant breeders

8 Alien substitutions and translocations

9 Protecting Yield Potential: The Role of Resistance toPathogens and Pests

11 Moving beyond Marginal Yields in Marginal Environments

11 Breeding for drought tolerance

14 Higher yielding durum wheats

15 Regional research on wheat for marginal environments

16 Improvements in Wheat Quality

17 Biotechnology and Wheat Improvement: An Example ofCollaboration

18 Crop and Natural Resource Management Research

19 Improved input use efficiency

19 Bed planting systems

20 Farmer participatory research

21 Information Management Tools for Sustainable Systems

22 Conclusions

24 A new research paradigm for new research impacts

24 The shape of things to come

27 References

iii

Acknowledgments

The authors are grateful to many colleagues within CIMMYT whocontributed information for this paper. Special thanks go to staff ofthe Wheat Program, as much of their research is described here,including A. Mujeeb-Kazi, I. Ortiz-Monasterio, J. Peña, W. Pfeiffer,M.P. Reynolds, R.P. Singh, K.D. Sayre, and P. Wall. L. Harringtonand J. White of the Natural Resources Group, and A. McNab andD. Poland of Information Services, also generously providedinformation for this paper. We thank the CIMMYT design sectionfor layout and production of the publication.

None of the work reported here would have been possiblewithout the continuining support of CIMMYT’s investors, themembers of the CGIAR. Amongst those we particularly thank ourcore investors.

iv

1

At the same time that we arewitnessing a proliferation ofagricultural innovations unlike anyseen previously, hunger and povertyremain the defining conditions of lifefor hundreds of millions of people.

New agricultural knowledge andtechnologies are announced almostdaily. The shifting alliances and theachievements of transnational seed-chemical-pharmaceutical companiesare minutely analyzed in the media. Itis easy to forget that this freneticactivity occurs in a sobering context—aworld of persisting hunger.

Even a small number of facts aresufficient to demonstrate the gravity ofthe world food situation. More than800 million people in developingcountries—20% of the population—cannot be certain that they will getenough to eat, because they lack theresources to grow or purchasesufficient food. The downward spiralof hunger and poverty remains seriousin many regions and countries. Anestimated 1.3 billion people live inhouseholds earning US$ 1 per day orless per person. Asia has 73% of theworld’s poor people (World Bank1997), and as we move into the newmillenium South Asia will continue tobe the home of half of the world’s poor.

Though Asia will have the highestabsolute number of poor people, thenumber of poor people in Sub-Saharan Africa (which currently has17% of the world’s poor) will grow by40% between 1990 and 2000, and thenumber of undernourished peoplewill rise by 70% between 1988-90 and2010 (World Bank 1997; Reeves,Pinstrup-Anderson, and Pandya-Lorch 1997).

Developing countries are projectedto increase their demand for cerealsby about 80% between 1999 and 2020(Pinstrup-Andersen and Pandya-Lorch 1997). Rosegrant et al. (1997)report that over the next two decadesglobal demand for wheat and maizecould rise by 40% and 47%,respectively. By 2020, it is expectedthat 67% of the world’s wheatconsumption and 57% of the world’smaize consumption will occur indeveloping countries.

Even if food crop productivityin developing countries remains atcurrent levels, by 2020 developingcountries will be importing 138million tons of wheat and 62million tons of maize every year. Inthese circumstances, how will weensure food security for the poorestof the poor?

New Wheats for a Secure,

Sustainable FutureTimothy G. Reeves, Sanjaya Rajaram, Maarten van Ginkel,Richard Trethowan, Hans-Joachim Braun, and Kelly Cassaday

2

Agriculture: An

Agent of Change

The role of more productive, profitablemaize and wheat systems in fosteringfood security, generating localemployment, raising local incomes,and thus alleviating poverty must notbe underestimated. A recent report(UNDP 1997) emphasizes thatagricultural research is the centralmeans of achieving those goals:“About three-quarters of the world’spoorest people live in rural areas,dependent on agricultural activitiesfor their livelihoods. For these people,pro-poor growth means raisingagricultural productivity, efficiency,and incomes.” The report points outwhy agriculture can succeed whereother initiatives might fail: “Raisingthe productivity of small-scaleagriculture does more than benefitfarmers. It also creates employment onthe farm and off—and reduces foodprices. The poor benefit most, becauseabout 70% of their consumption isfood, mostly staples, and regularsupplies and stable prices can greatlyreduce the vulnerability of the poor.Strong support to small-scaleagriculture was at the core of the mostsuccessful cases of povertyreduction—such as China in 1978-85,Malaysia since 1971, and India in theearly 1980s.”

In these circumstances, thechallenges for research—and theopportunities to alleviate muchhuman suffering—are clear. We willhave to develop the innovations thatmake it possible for people to benefitfrom more efficient, low-cost systemsfor food production. These systems

must function without mining thenatural resources on which agriculturedepends. They are needed urgently infavored as well as less favoredagricultural areas.

In this paper, we review strategiesused by the International Maize andWheat Improvement Center(CIMMYT) and its partners to developsustainable wheat1 productionsystems for favored and marginalareas. These strategies aim to achievean optimal combination of the bestgenotypes (G), in the rightenvironments (E), under appropriatecrop management (M), andappropriate to the needs of the people(P) who must implement and managethem (Reeves 1998, 1999). Eachvariable in this GxExMxP“sustainability equation” is addressedin the sections that follow. After furtherdefining what we mean by“sustainable technology,” we:

• Review new options for raising wheatyield potential.

• Discuss research on disease and stresstolerance, which is aimed atprotecting yield potential in farmers’fields. We give special emphasis todrought tolerance.

• Describe advances in durum wheatyield potential which may proveparticularly valuable in marginalenvironments.

1 In this paper we focus on strategiesrelated to wheat, although CIMMYT’sresearch mandate encompasses maize aswell. We also give greater attention towheat genetic improvement than tocrop and natural resource managementresearch, but readers should be advisedthat CIMMYT engages in a great deal ofcrop and resource managementresearch, for wheat as well as maize. Fora general overview, see our annualreport, CIMMYT in 1998-99: Science toSustain People and the Environment.

3

• Provide an overview of other wheatresearch initiatives for marginalenvironments.

• Review the role of biotechnology inwheat improvement.

• Describe recent research on wheatquality. For many poor farmers, anincrease in wheat quality means acorresponding increase in income.

• Briefly review recent initiatives incrop and natural resourcemanagement research in wheat.

We conclude by summarizing thelatest data on the global impacts ofour wheat research and by discussingtrends that could affect whether andhow this impact is maintained intothe future.

Prerequisites

for Sustainable

Agriculture

To be sustainable, farming systemsmust be biologically sensible,economically viable, environmentallysound, socially acceptable, andpolitically supportable (Reeves 1998,1999):

• Sustainable farming systems must bebiologically sensible. For example,the choice of crop(s), theirmanagement, and the level ofintensification must be consistentwith the biophysical realities of thefarming system.

• Sustainable farming systems must beeconomically viable at the farm andnational levels. Poor farmers cannot

invest in systems that will notproduce reasonable yields and (evenbetter) cash income, now and in thefuture. At the national level, thereality in most developing countriesis that economic well-being anddevelopment are almost invariablybased on productive and profitableagriculture, the “engine room” ofsubsequent industrialization.

• Sustainable farming systems mustbe environmentally sound.Economic success in agriculturecannot come at the expense of oursoils, air, water, landscapes, andindigenous flora and fauna.

• Sustainable farming systems mustbe socially acceptable. They must beappropriate to the people who,relying on their own meagerresources, are responsible forimplementing and managing them.The need for socially acceptablesystems implies the need for a betterunderstanding of farmer andcommunity needs and values, aswell as better targeting oftechnology to meet local conditions.

• Finally, sustainable farming systemsmust be politically supportable.Political support depends largely onsuccessfully meeting the first threerequirements of sustainability. Ifeconomic growth is catalyzed byagriculture within anenvironmentally sound, sociallyacceptable framework, politicianswill continue to view agriculture asjustifying support.

All of these components combineto form the whole: sustainableagriculture. If one is neglected, it canseriously reduce the rate and extent ofprogress towards sustainability andfood security.

4

Breeding Wheats

for Lasting Food

Security

CIMMYT’s wheat breedingmethodology is tailored to developwidely adapted, disease resistantgermplasm with high and stableyield across a wide range ofenvironments—favorable as well asmarginal. To focus this work, wehave grouped wheat productionareas in developing countries into 12“mega-environments.” A mega-environment is a broad but notnecessarily contiguous geographicalarea, usually international andfrequently transcontinental. Mega-environments are defined in terms ofthe type of wheat cultivated (spring,facultative, or winter wheat), theamount of water available to thecrop, temperature regime, mineraltoxicity in the soil, and the majordiseases and pests that limit foodproduction.

CIMMYT wheat breeders, throughcollaboration with national wheatresearch programs and genebanks,scour the world for new and differentsources of yield potential and othertraits of interest. We give the utmostattention to genetic diversity withinCIMMYT germplasm to minimize therisk of genetic vulnerability, since ourbreeding materials are used in researchprograms worldwide, and thenumerous varieties developed fromthose breeding materials are grown byhundreds of millions of farmers. Wealso believe that the use of geneticallydiverse material is mandatory forfuture increases in yield potential andyield stability. At CIMMYT, parental

groups of lines for crossing in anyyear consist of 500-800 lines. Twice ayear around 30% of parental stocksare replaced with outstandingintroductions. In addition, commercialcultivars from national agriculturalresearch systems (NARSs) and non-conventional sources (e.g., durumwheat and alien species) are used toincorporate desired traits byrecombination or translocation. Theintroductions are mostly used as thefemale parent to preserve cytoplasmicdiversity.

Options for

Increasing Yield

Potential

Like most wheat improvementprograms, the CIMMYT wheatimprovement program has manyreasons for seeking to raise—andprotect—genetic yield potential. Highyield potential, assessed in breeders’trials, is positively associated withsuperior crop performance in farmers’fields, even in stressed environments.Another consideration is that mostfarmers readily adopt and shareimproved wheat seed, even in areaswhere problems with infrastructureand lack of farmer support servicesfrustrate the adoption of otheragricultural inputs and practices.

Those may be regarded as the“humanitarian” reasons for seekinghigher yields, but it is important toremember that there are alsocompelling environmental reasons tobreak yield barriers. We must berealistic about changing land use

5

patterns and their implications foragriculture. There is limited scope toopen new land for crop production, andthere is an even more urgent need toprotect land (in particular, marginalland) from inappropriate uses. In recentdecades developing countries havefortunately relied more on increasedyields than on an expansion of croppedarea to feed their populations. Between1961 and 1990, yield increases accountedfor 92% of the additional cerealproduction in the developing world(Reeves, Pinstrup-Anderson, andPandya-Lorch 1997). When farmers instable, high production environmentsobtain better yields, the need to intensifyproduction in fragile agriculturalsystems is reduced, offering a muchmore sustainable approach to meetinglong-term demand for cereal productionin developing countries.2 Becausehigher yielding lines are frequently bredto use inputs such as nutrients andwater more efficiently, higher yields arenot obtained at a higher cost to theenvironment. As our CIMMYTcolleague, Nobel Laureate NormanBorlaug, has said, “The only way foragriculture to keep pace with populationand alleviate world hunger is to increasethe intensity of production in thoseecosystems that lend themselves tosustainable intensification, whiledecreasing intensity of production in themore fragile ecologies” (Borlaug andDowswell 1997).

The selection of segregatingpopulations and consequent yieldtesting of advanced lines is paramountfor identifying high yielding, inputresponsive wheat genotypes. Theincrease in yield potential of CIMMYTcultivars developed since the 1960s isshown in Figure 1 (K. Sayre, pers.comm.). The data do not indicate thatwe are approaching a yield plateau,and the performance of recentlyreleased lines such as Attila andBaviacora, and of Lr19-derived Veery,indicates that yield potential has beenfurther enhanced.

With yield, a complex trait stillnot well understood genetically orphysiologically, the use of proven,high yielding sources, as well asgenetically diverse germplasm, willcontinue to be paramount forincreasing yield potential. Geneticdiversity and the opportunity for itsrecombination through crossing willbe important to break undesiredlinkages and increase the frequency

2 For example, if India were suddenlyrequired to produce its current wheatharvest using the technologies of 30years ago, Indian farmers would have tobring more than 40 million hectares ofadditional land into production. Thewheat varieties developed in the pastthree decades were instrumental inpreventing damage to areas that are notwell suited to agriculture.

10,000

9,600

9,200

8,800

8,4001965 70 75 80 85 90 95

Variety year of release

Grain yield kg/ha at 12% H20

Figure 1. Grain yield trend for semidwarf breadwheat lines developed at CIMMYT since 1966,under conventional planting, average for 1997,1998, and 1999 crop cycles at CIANO, Cd.Obregón, Mexico.Source: K.D. Sayre, CIMMYT.

Yield = -6.22*+104+36.05x(kg/ha/year)r2 = 0.772Annual yield increase = 0.39%/year

6

of desirable alleles. Futurebreakthroughs in yield potential arelikely to come from such geneticallydiverse crosses. Examples are givenbelow, along with a description ofother efforts to raise both spring andwinter wheat yield potential.

Gene Pools of Winter andSpring Hexaploid WheatsThe variability currently availableamong spring and winter hexaploidwheats is still extensive. New highyielding sources from within theCIMMYT Bread Wheat Program andfrom around the world are identifiedand intercrossed. For example, highyielding spring wheat lines from SouthAsia and China are regularlyintercrossed with the highest yieldinglines identified in Mexico, followed byselection for types superior to eitherparent, carrying all desirable genes.Likewise elite winter wheats areintercrossed. Considerable progresscan still be made in this way as yield iscontrolled by many genes and theoptimal combinations of these genesfor any particular environment maynot yet have been realized.

Introgressing Spring andWinter Wheat Gene PoolsBy introgressing genetic variabilityfrom winter wheats, breeders haveconsiderably augmented the yieldpotential of spring wheats. The Veerywheats, developed from crosses ofCIMMYT spring wheats and Russianwinter wheat, represented a quantumleap in spring wheat yield and wideadaptation during the 1970s and 1980s(CIMMYT 1986) (their contribution todrought tolerance is discussed later).More recently, the spring bread wheat

Attila, developed from crosses withwestern European and US winterwheats, has rapidly gained ground onthe Indian subcontinent. New evidenceindicates that yield potential in winterwheat may also benefit from crosseswith high yielding spring wheats.

Chinese Wheats: AWellspring of DiversityBefore the mid-1980s, only a limitedamount of wheat germplasm fromoutside China was available toChinese breeders. Since the mid-1980s, CIMMYT and Chinesescientists have worked together tobenefit from the diversity in eachothers’ wheat germplasm. Morethan 100 Chinese varieties containCIMMYT germplasm, and up to20% of new CIMMYT spring wheatshave Chinese wheats in theirpedigrees. Apart from its resistanceto biotic pests such as scab andKarnal bunt, modern Chinesegermplasm offers new alternativesfor raising the yield potential ofwheat; yields of elite Chinesewheats in China can exceed 10 t/ha.

Hybrid WheatsThe expression of heterosis for yield inwheat can be high. Although it hasbeen well documented, heterosis hasnot been exploited commercially toany great extent. Hybrids offer theunique opportunity of combiningdifferent gene pools in the productionof the F1 hybrid. Because heterosis is,to some extent, a function of geneticdistance, CIMMYT is well positionedto exploit this need for geneticdiversity. During the past three years,CIMMYT hybrids have producedyields that are 15-20% higher thanthose of commercially grown cultivars

7

in Mexico, and levels of heterosis ofa similar maximum size have beenreported. The difficulty of producingF1 seed in a cost-effective wayremains the greatest limitation to theexploitation of such hybrids, butCIMMYT breeders expect to resolvethis issue by introgressingoutcrossing traits.

LandracesMany high yielding CIMMYTwheats have a considerable numberof landraces in their pedigrees. Acoefficient of parentage analysisreveals that on average CIMMYTadvanced lines contain as many as50 landraces in their genetic history.Breeding programs have still notexploited all of the yield-controllinggenes available in landraces.Landraces may also provide novelsources of adaptation, which willallow breeders to select more stable,high yielding lines. As yieldsincrease, consumer preferences willalso turn to increased quality andtaste. Here, locally preferredlandraces can play a very new andexciting role.

Improved Plant IdeotypeCIMMYT breeders are usingincreased knowledge of thephysiological bases of yield to definea range of optimal wheat plantideotypes. We are examining plantswith large spikes, which containmany grains per spikelet (see photo,inside front cover). The optimizationof source-sink relationships is alsobeing examined with a view toobtaining a better balance of grain-filling characters. The hexaploidwheat and other gene pools arebeing searched for examples of

extreme expression of thesecharacters. We believe advances ofyield potential on the order of at least20% in optimum conditions can stillbe realized by fine-tuning the source-sink relationships in wheat.

Phenological TraitsBy manipulating photoperiod andvernalization genes, we areattempting to tailor genotypes tospecific environments. Photoperiodand vernalization genes optimize thetiming and duration of flowering andgrain-filling, thereby influencing thewheat plant’s eventual yield. Newand different sources of these genesare being exploited through the use ofhigh-latitude germplasm from CentralAsia and Canada.

Physiological TraitsA strong body of evidence nowindicates that physiological traits maycomplement early-generationphenotypic selection in wheat. Geneticprogress in increasing yield potentialis closely associated with increasedphotosynthetic activity (Rees et al.1993). Photosynthetic activity as wellas yield potential have increased overthe past 30 years by some 25%. Thesefindings may have major implicationsfor CIMMYT’s future selectionstrategy, since there is evidence thatwheat genotypes with higherphotosynthesis rates have lowercanopy temperatures, a characteristicthat can be measured easily, quickly,and cheaply. Canopy temperaturedepression (CTD) is the cooling effectexhibited by a leaf as transpirationoccurs. Canopy temperaturedepression and stomatal conductance,measured on sunny days during grainfilling, have shown a strong

8

association with yields of semidwarfwheats grown under irrigation, in bothtemperate (Fischer et al. 1998) andsubtropical environments (Reynolds etal. 1994). In addition, CTD measuredon large numbers of advancedbreeding lines in irrigated yield trialswas a powerful predictor ofperformance, not only at the selectionsite but also for yield averaged across15 international sites. Canopytemperature depression has beenshown to be associated with yielddifferences between homozygous linesin warm environments, indicating apotential for genetic yield gains underconditions of heat in response toselection for CTD (Reynolds at al.1998). Breeders have found CTD to behighly correlated with yield underheat conditions among elite lines (vanGinkel and Trethowan, unpublished),and the technique may be particularlyuseful for more efficiently selectingwheat genotypes adapted toenvironments where heat is aproduction constraint.

Synthetic Wheats:Delivering Diversity toPlant BreedersSynthetic wheats are the result of across between two relatives of putativeprogenitors of wheat, Triticumturgidum and T. tauschii, withsubsequent chromosome doubling.Historically (10,000 to 8,000 years ago),this cross has probably occurred ononly a few occasions. As a result, thegenetic resources of these two specieshave been sampled in only a limitedway in the development of breadwheat. CIMMYT holds a large numberof T. tauschii accessions from which

many new synthetic hexaploidwheats have been made (about 650 todate). These synthetics possess arange of positive traits, includingresistance to such diseases as Karnalbunt, fusarium head scab, andhelminthosporium leaf blotch, andtolerance to heat, drought,waterlogging, and late frost atflowering. They are spring types thatare highly crossable to advancedbread wheats, which means that theymay be used easily in breedingprograms. Through this approach,CIMMYT breeders have not onlybeen able to take advantage of thenew variation from T. tauschii, buthave also found a new way tointrogress traits from elite durumwheats into bread wheat. Syntheticsor their derivatives may also proveuseful in the production of hybridwheat and the improvement of breadwheat quality.

Alien Substitutionsand TranslocationsThe 1B/1R translocation (discussedalso in the section on droughttolerance) led to a revolution in thebroad adaptation of wheat. Thistranslocation from rye increasedwheat biomass, harvest index, and—especially—wide adaptation, whichspurred improvements in wheat yieldin most spring wheat environments.More recently, a translocated segmentfrom Agropyron sp. containing the leafrust resistance gene Lr19 has beenlinked with a 5-10% increase in yieldin adapted backgrounds. Other aliensources of higher yield are alsocurrently under evaluation.

9

Protecting

Yield Potential:

The Role of

Resistance to

Pathogens and Pests

Over the past few decades, the gainsfrom breeding for disease resistanceare likely to have been at least asimportant as the gains from breedingfor increased yield potential (Byerleeand Moya 1993). A recent survey ofwheat breeders in developingcountries indicated that among thetypes of materials used in crossing(including the breeder’s own advancedlines, advanced lines obtained fromother countries, wild relatives, andlandraces), materials from CIMMYTinternational nurseries are the mostfrequently crossed in pursuit of diseaseresistance goals (Rejesus, van Ginkel,and Smale 1996).

CIMMYT’s global effort to breedwheats with diverse and durableresistance will protect global foodsecurity by reducing the incidence ofdisease epidemics. It will also protectthe environment and farmers’ incomes,by reducing dependence on pesticidesfor disease and pest control. InCIMMYT’s target mega-environments,important fungal diseases of wheatcaused by obligate parasites includethe rusts (one or more of which are themost economically important diseasesin most wheat productionenvironments), powdery mildew, andthe bunts and smuts. Widespread

diseases caused by facultative fungalparasites include septoria tritici blotch,septoria nodorum blotch, spot blotch,tan spot, head scab, and a suite of rootrots.

The obligate parasites are highlyspecialized, and significant variationexists in the pathogen population forvirulence to specific resistance genes.The evolution of new virulence (races)through migration, mutation, andrecombination of existing virulencesand their selection is more frequent inrust and powdery mildew fungi. Forthis reason, these diseases haverequired constant vigilance andattention from breeders. Physiologicalraces are also known to occur for mostbunts and smuts, although evolutionand selection of new races is lessfrequent. Because most bunts andsmuts are easily controlled by chemicalseed treatment, little effort is currentlyplaced on breeding for resistance,except for resistance to Karnal bunt.Successful changes in pathogen racesare even less frequent in the facultativeparasites mentioned earlier.

Since wheat cultivars derived fromCIMMYT germplasm are grown over alarge area and are exposed to a varietyof pathogens under conditions thatmay favor disease development, ourstrategy has been to utilize resistancesources that are as diverse as possibleand have shown durability. Geneticdiversity and durability of resistanceagainst diseases caused by pathogenssuch as the rust pathogens are vital forlong-term food security. Resistancescaused by race-specific genes becomeineffective in a short time (in five yearson average at the global level and inthree years for leaf rust, Pucciniarecondita, in Mexico). In contrast,

10

cultivars with durable resistance haveshown stable resistance for over 50years at the global level. Consider theresistance to stem rust (P. graminis).McFadden in the US transferred theSr2 gene complex from a tetraploidemmer wheat to hexaploid breadwheat in the 1920s (McFadden 1930).Borlaug in Mexico used this source ofresistance in his breeding program inthe 1940s, and since then this gene, inconcert with several known andunknown major and minor genes, hasformed the basis of durable resistanceto stem rust in CIMMYT wheatgermplasm.

Following the lesson learnt fromstem rust research, CIMMYT’s wheatbreeding in the last three decades hasfocused on utilizing diverse sources ofslow rusting resistance to P. reconditaand yellow rust (P. striiformis). Geneticanalyses of durable resistance indicatethat effective disease control can beachieved by combining from three tofive minor, slow rusting genes in asingle cultivar. Such resistance isexpected to provide sufficientprotection to farmers’ crop against allbiotypes over a long period. Currentlywe are also attempting to identifymolecular markers for each of the slowrusting genes present in CIMMYTwheats. If this strategy is successful,breeding programs will be able toincorporate known combinations ofminor genes, develop a global strategyfor their deployment, and at the sametime enhance genetic diversity infarmers’ fields.

Recent analysis of trials conductedin northwestern Mexico confirms thatprogress in protecting yield potentialthrough genetic resistance to leaf rustis about three times as great asadvances in yield potential itself (R.P.Singh and K.D. Sayre, pers. comm.).The economic benefits of CIMMYT’sstrategy of incorporating non-specific,durable resistance to leaf rust intomodern bread wheats have beenestimated using data on resistancegenes identified in cultivars, trial data,and area sown to cultivars innorthwestern Mexico. Even under themost conservative scenario, the grossbenefits generated in this region onabout 120,000 ha of wheat from 1970 to1990 were US$ 17 million (in 1994 realterms) (Smale et al. 1998). At the globallevel, where a considerable area issown to cultivars carrying non-specificresistance, the benefits must becorrespondingly large.

Resistance to the diseases causedby facultative parasites, such asSeptoria tritici and Fusariumgraminearum, also involves genes thathave additive effects. Tremendousprogress has been made at CIMMYT indeveloping semidwarf wheats thathave adequate resistance to Septoriatritici. Sources contributing toresistance include wheats from France,Brazil, China, and Russia. Morerecently we have identified syntheticwheats (T. turgidum x T. tauschii)possessing good resistance to septoriatritici blotch. This new geneticdiversity is currently being transferredto CIMMYT wheats.

11

Moving beyond

Marginal Yields in

Marginal

Environments

Limited water availability is probablythe most common stress that affectsfarmers in marginal environments, butthey also have to contend with factorssuch as diseases, acid soils, extremecold and heat, waterlogging, andmineral deficiencies and toxicities. Aregion is defined as marginal whenwheat production drops to 70% ofoptimal yield levels, as in, for example,the highland areas from Turkey toAfghanistan, the dryland areas of WestAsia and North Africa (WANA), muchof Ethiopia, and the dryland areas ofcentral and southern India (Table 1).3

Our discussion of CIMMYT’sresearch directed at marginal areasbegins with a review of the methodsused in breeding drought tolerantwheats. Next, we describeachievements in durum wheatbreeding, given the considerableamount of durum wheat grown inmarginal areas. We conclude with anoverview of specific researchinitiatives in regions where marginalenvironments present a series ofchallenges to wheat production. As thefollowing sections indicate, CIMMYT

researchers and their collaboratorsare implementing a combination ofstrategies to ensure that farmers inmarginal areas are no longer destinedto obtain marginal yields.

Breeding for DroughtToleranceThe annual gain in genetic yieldpotential in drought environments isonly about half (0.3-0.5%) of thatobtained in irrigated, optimumconditions. Many investigators haveattempted to produce wheat adaptedto semiarid environments but withlimited success. The CIMMYT WheatProgram follows a system ofbreeding for drought tolerance inwhich yield responsiveness iscombined with adaptation to droughtconditions. Because most semiaridenvironments differ significantly inannual precipitation distribution, andbecause water availability also differsacross years in these environments, itis prudent to construct a geneticsystem in which plant responsivenessprovides a bonus whenever higherrainfall improves the productionenvironment. With such a system,

Table 1. Portions of wheat producing regions ofthe world that are defined as marginal

Total wheat PercentRegion area (000 ha) marginal

West Asia/North Africa 28,300 65Central Asia and the Caucasus 15,000 80South Asia (Subcontinent) 34,500 35East Asia (including China) 30,100 13Eastern Africa 1,500 27Southern Africa 1,300 91Southern Cone of South America 7,400 60Andean Region of South America 300 18Mexico/Central America 900 43Total 119,300 45

3 Note that, although improved varietieshave a role to play in these areas,considerable gains will also result fromimproved crop and resource management,especially measures to conserve andutilize moisture more efficiently in rainfedareas. Some of these practices arediscussed later in this publication.

12

improved moisture is immediatelytranslated into greater yield gains forfarmers.

Why do we believe that this can bedone? One compelling piece ofevidence comes in the form of Veery S,which combines high yieldperformance in favorable environmentsand adaptation to drought in moremarginal areas. When Veery S wastested in 73 environments in the early1980s, its performance differed fromthat of other high yielding varieties. Ityielded better than other cultivars notonly in high yielding environments butalso under reduced irrigation (Table 2).What made this line different was thatit carried the 1B/1R translocation fromrye. By 1990, 63% of the dryland wheatarea in developing countries was sown

to semidwarf wheats (Byerlee and Moya1993). Many of these wheats possessedthe 1B/1R translocation, which hadbeen incorporated into hundreds ofgenetically different backgrounds andmade available to breeders throughoutthe world.

We have conducted severalexperiments to compare theperformance of the newest and mostwidely adapted wheat germplasm to theperformance of commercial cultivarsfrom countries in three marginal, lowrainfall mega-environments, underconditions simulating thoseenvironments (Calhoun et al. 1994; vanGinkel et al. 1998; Tables 3 and 4). Themost widely adapted CIMMYT linesyielded better than the commercialcultivars in all of the simulatedenvironments. Recent adoption of

Table 3. Wheat genotypes representing adaptation to different moisture environments

Mega-environment (ME) Genotype

ME1 (Irrigated environment) Super Kauz, Pavon 76, Genaro 81, Opata 85ME4A (Mediterranean Region) Almansor, Nesser, Sitta, Siete CerrosME4B (Southern Cone, S. America) Cruz Alta, Prointa Don Alberto, LAP1376,

PSN/BOW CM69560ME4C (South Asian Subcontinent) C306, Sonalika, Punjab 81, Barani

Source: Calhoun et al. (1994).

Table 2. Effect of the 1BL/1RS translocation on yield characteristics of 28 random F2-derived F6 lines from the cross Nacozari/Seri 82 under reduced irrigation

MeanCharacteristic 1BL/1RS 1B difference

Grain yield (kg/ha) 4,945 4,743 202 *Above-ground biomass at maturity (t/ha) 12,600 12,100 500 *Grains/m2 14,074 13,922 152 NSGrains/spike 43.5 40.6 2.9 *1,000-grain weight (g) 37.1 36.5 0.5 *

Source: Villareal et al. (1995).Note: NS = not significant; * = significant at the 0.05 level.

13

CIMMYT germplasm in thoseenvironments supports the model ofcombining input efficiency and inputresponsiveness.

Another piece of evidence isNesser, an advanced line withsuperior performance in droughtconditions. Nesser was bred atCIMMYT-Mexico and identified bythe CIMMYT Mediterranean programlocated at the International Center forAgricultural Research in the DryAreas (ICARDA) in Syria. The crosscombines the high yielding CIMMYTvariety Jupateco 75 and the droughttolerant Australian variety W3918A.The performance of Nesser in thedryland environments of WANA hasbeen widely publicized (ICARDA1993), and the line is considered torepresent a uniquely drought-tolerantgenotype. This line was selected atCIMMYT/Mexico under favorableconditions, and it carries acombination of input efficiency andhigh yield responsiveness. In theabsence of rust, its performance isquite similar to that of Veery S.

A breeding scheme to achieve thecombination of yield responsivenessand drought tolerance in wheat ispresented in Table 5. This method issupported by research on wheat aswell as other crops, in which testingand selecting in a range ofenvironments, including well-irrigatedones, has identified superiorgenotypes for stressed conditions (see,for example, Ehdaie, Waines, and Hall1988; Duvick 1990, 1992; Bramel-Coxet al. 1991; Uddin, Carver, and Clutter1992; Zavala-Garcia et al. 1992; andCooper, Byth, and Woodruff 1994). Theapproach results in the selection ofgermplasm that is adopted by farmersbecause it translates improvedenvironmental conditions into yieldgains. The traditional methodology ofselecting only under droughtconditions and narrowly relying onlandrace genotypes does not moveyield levels significantly beyond thoseusually obtained, and it does notprovide the farmer with a bonus inyears when rainfall is higher.

Table 4. Grain yields (kg/ha) of selected wheat genotypes grouped by adaptation andtested under moisture regimes in the Yaqui Valley, Mexico, 1989/90 and 1990/91

Full Late Early ResidualAdaptation group irrigationa droughtb droughtc moistured

ME1 (Irrigated environment) 6,636 a 4,198 a 4,576 a 3,032 aME4A (Mediterranean Region) 6,342 b 3,990 ab 4,390 b 3,032 aME4B (Southern Cone, S. America) 5,028 c 3,148 bc 4,224 b 2,359 cME4C (South Asian Subcontinent) 4,778 c 3,245 bc 3,657 c 2,704 b

Source: Calhoun et al. (1994).Note: Means in the same column followed by the same letter are not significantlydifferent at P=0.05.a Received 5 irrigations.b Received 2 irrigations early, before heading.c Received 1 irrigation for germination and 2 post-heading.d Received 1 irrigation for germination only.

14

Higher YieldingDurum WheatsAlthough durum wheat is not cultivatedas widely as bread wheat, it occupies aspecial niche in the developing world.Durum wheat is generally sown inmarginal environments subject to greatclimatic fluctuations during the growingseason. The durum crop may experienceheat and drought at different timesduring its growth cycle. Most of thedeveloping world durum area isconcentrated in the countries of WANA,but durum is also grown in central andsouth India,4 Ethiopia, Mexico,Argentina, Peru, Kazakhstan,

Azerbaijan, and Ukraine. Often thecrop is grown by poor people who relyon it for a high proportion of caloriesin the diet or for income—as durum insome areas fetches a premium in thelocal market.

Short-cycle, semidwarf durumwheat varieties recently tested innorthwestern Mexico produced aremarkable 89 kg of grain per hectareper day, for a final tally of 11.7 t/ha atharvest (W. Pfeiffer, pers. comm.). This

4 In parts of India, durum production isrelegated to the hottest and driestenvironments.

Table 5. Methodology for breeding drought-tolerant wheat that is also responsive tofavorable environmental conditions

Generation Activity

F1 Crosses involving widely adapted germplasm, representing yieldpotential, yield stability, and input responsiveness, with lines carryingproven drought tolerance in the setting of the respective drought mega-environments (ME4A, ME4B, or ME4C), and input (water) use efficiency.Winter wheat and synthetics are emphasized.

F2 Individual plants are raised under irrigated and optimally fertilizedconditions, and inoculated with a wide spectrum of rust virulence. Onlyrobust and horizontally resistant plants are selected. These plants mayrepresent adaptation and responsiveness to favorable environmentalconditions.

F3 The selected F2 plants are evaluated as F3s in a modified pedigree/bulkbreeding system (Rajaram and van Ginkel 1995) under rainfed conditionsor very low water availability. The selection is based on such criteria asspike density, biomass/vigor, and grains/m2, among others (van Ginkelet al. 1998). This index helps identify lines that may adapt to conditions inwhich water is limited (that is, lines that are input efficient).

F4 Selected lines from F3 are further evaluated under optimum conditions,as for the F2.

F5 As F3.

F6 As F4.

F7, F8 Simultaneous evaluations under low and intermediate (representing thehigher rainfall years in marginal drought environments) water regimes.Selection of those lines showing outstanding performance under bothconditions. Further evaluation in international environments is carriedout for verification.

15

is an increase of more than 20% overthe previous generation of durumwheats. Generally average yields ofdurum wheat in farmers’ fields innorthwestern Mexico are 6 t/ha, andthe world average is 2-3 t/ha. If theserecently tested wheats retain some oftheir yield advantage in marginalconditions, they may prove to be avaluable asset for breeding programs.

Regional Researchon Wheat for MarginalEnvironmentsWest Asia and North Africa. Aboutone-third of the area planted to wheatin the developing world is located inmarginal environments plagued bydrought and soil problems. Theseproblems are frequently exacerbatedby a lack of infrastructure and farmersupport services. Most of the world’sdrought-prone wheat area isconcentrated in the WANA region(Table 1). Wheat is the principal foodsource for people in WANA, who onaverage consume more than 145 kg/cap/yr, one of the highest levels of percapita consumption in the world.

CIMMYT efforts aimed atimproving wheat production inWANA are conducted in conjunctionwith ICARDA. The CIMMYT/ICARDA Joint Dryland WheatProgram for West Asia and NorthAfrica seeks to increase wheatproductivity by developing springbread and durum wheats that arebetter adapted to the WANA region.Wheats developed or identified by theprogram are widely adapted andpossess enhanced disease and insectresistance, as well as better tolerance tothe prevalent abiotic stresses. This iswhy our partners in the regionincreasingly select them for use in their

own breeding programs. Farmeradoption of CIMMYT- and CIMMYT/ICARDA-derived varieties in WANAcontinues to increase, with more than90 wheat varieties released in 21countries in the region over the past 10years.

The Turkey/CIMMYT/ICARDAInternational Winter WheatImprovement Program (IWWIP) basedin Ankara, Turkey, came into existence11 years ago with the purpose ofgenerating winter wheats fordeveloping countries, particularly inthe WANA region. Over the past twoyears, IWWIP has expanded itscollaboration with winter wheatprograms in the developing world.New research partnerships withcolleagues from Central Asia and theCaucasus have greatly increased thenumber of cooperators.

The program is devoting particularattention to improving resistance toyellow rust, which is the most seriouswinter wheat disease in WANA. Itconducts trials using artificialinoculation in Ankara, Konya, andEskisehir (Turkey), Aleppo (Syria), andIran. It is also conducting research onmicronutrients aimed at identifyingzinc-efficient wheats to be used incrosses and alien materials that may bepotential sources of zinc efficiency. Atpresent, rye and triticale seem to be thebest sources, but other alien species arealso being tested at Turkey’s ÇukurovaUniversity.

Central Asia and the Caucasus. Therepublics of Central Asia and theCaucasus are relatively diverse inclimate, agricultural production, andpopulation. What these eight countries(Armenia, Azerbaijan, Georgia,Kazakhstan, Kyrgyzstan, Tadjikistan,Turkmenistan, and Uzbekistan) have in

16

common is that they are all in transitionfrom being centrally planned economiesto becoming market-oriented ones.Nearly 15 million hectares are plantedto wheat in the region, but with theexception of Kazakhstan, all countrieshave to import wheat to satisfydomestic demand. A major objective oftheir governments is to become self-sufficient in wheat.

In 1992, after the political situationchanged, CIMMYT re-establishedcontacts with research programs in theregion. In 1998, CIMMYT wasmandated by the CGIAR to address theneeds for wheat germplasm in thisregion. Breeders and researchadministrators from the region havevisited IWWIP in Ankara or CIMMYTin Mexico, and CIMMYT scientists havevisited several of the newlyindependent nations. In 1998 CIMMYTopened a regional office in Kazakhstan.There is now active exchange ofgermplasm and information. In thefuture, CIMMYT will initiate shuttlebreeding programs with the region. Ajoint CIMMYT/Kazakhstan breedingprogram to combine quality, droughttolerance, and disease resistance in highlatitude spring wheat is in the planningstages. If successful, it would contributeto the food security not only ofKazakhstan, but of the whole region.

Eastern Africa. Now in its fourth phase,the CIMMYT/Canadian InternationalDevelopment Agency Eastern AfricaCereals Program (EACP) has as its mainobjective to increase maize and wheatproduction and productivity in easternAfrica. During its third phase, thewheat component of the programfocused heavily on developingsustainable production systems for themajor wheat growing environments inthe region and on strengthening

national program commitment andcapacity for long-term experimentation.During 1993-96, Kenya, Ethiopia, andUganda released 13 CIMMYT-relatedbread wheat and durum wheatvarieties. Studies conducted by theEACP in collaboration with Ethiopiaand Kenya found that reduced or zerotillage produced either the same orbetter yields than conventional tillagesystems. The EACP also developedagronomic recommendations toimprove yields and nitrogen useefficiency in areas that experiencewaterlogging problems. Anencouraging fact brought to light in arecent report by the EACP is thatseveral decades of breeding durum andbread wheats from CIMMYTsemidwarf wheats in Ethiopia haveresulted in annual increases of 1.5-2.0%in yield potential based on rainfedexperiments.

Improvements in

Wheat Quality

Often wheat quality is perceived to beimportant only to large-scale farmersdedicated to commercial production. Infact, traits related to quality in wheatare even more important for many poorfarmers, whose incomes may increase ifthey can produce wheat that receives aprice premium for its qualitycharacteristics.

Several studies have concluded thatwild diploid species carrying A- and D-genomes have greater allelic variationthan cultivated wheat for gene locicontrolling glutenin subunits (Wainesand Payne 1987; Lagudah and Halloran

17

1988; Ciaffi et al. 1992; Lafiandra,Ciaffi, and Benedetelli 1993; William,Peña, and Mujeeb-Kazi et al. 1993).These alien genes offer a potentialmeans of expanding the number ofallelic variants controlling proteinswith desirable quality effects in wheat.Several of the synthetic hexaploidsdeveloped from accessions of diploidTriticeae species (T. tauschii, T.boeoticum, T. monococcum, and T. urartu)and durum wheat have been examinedin relation to grain characteristicsassociated with end-use quality ofbread and durum wheats. The analysesrevealed that T. tauschii may be usedfor substantially increasing the numberof high molecular weight glutenin(HMWG) subunits present in breadwheat (HMWG subunit composition isimplicated in the definition of glutenstrength in both bread wheat anddurum wheat) (Payne et al. 1981;Pogna et al. 1990).

We have also examined variabilityfor quality (grain hardness, proteincontent, and SDS-sedimentation) aswell as the relationship betweenquality and HMWG and low molecularweight glutenin (LMWG) subunitcomposition (SDS-PAGE) in 137accessions of T. dicoccon. Resultsconfirm previous findings that T.dicoccon has more diverse geneticvariability for alleles involved in thesynthesis of gluten-type proteins thancultivated wheat. T. dicoccon should beconsidered a good potential source forimproving gluten strength in breadand durum wheat.

In the past three years, thefrequency of high quality CIMMYTbread wheats has increaseddramatically. A modification of thecrossing strategy, emphasizing highquality parents, was implemented in

the early 1990s. Quality testing ofadvanced generation breedingmaterials was increased over the pastfew years. Now these two strategieshave come to fruition. In the nearfuture, about 75% of CIMMYT’s newbread wheat germplasm will becompetitive for quality standards inthe marketplace.

Biotechnology

and Wheat

Improvement: An

Example of

Collaboration

By drawing on the power ofbiotechnology, CIMMYT seeks to makeplant breeding more efficient and, insome cases, to improve wheat in waysthat have eluded conventionalbreeding approaches. The comparativegenetic mapping of cereal genomes hasidentified a vast amount of conservedlinearity of gene order (Devos andGale 1997). This observation is likely toaccelerate the application ofquantitative trait loci (QTL) in wheat,as well as aid in the identification ofgenes required for introgression fromalien species. Given the low number ofloci tagged at present in wheat, theproblems related to developing a high-density map for wheat (Snape 1998),and the limited progress to identifyQTL for yield in wheat, we believe thatthe impact from this linearity on wheatimprovement will be significant.

18

An extremely positive developmentin CIMMYT’s efforts to applybiotechnology to wheat improvementis participation as a core partner in theCooperative Research Centre (CRC) forMolecular Plant Breeding, establishedand supported under the Australiangovernment’s Cooperative ResearchCentres Program. The CRCcollaboration features two mainprojects. The first projects aims toidentify molecular markers forresistance to leaf rust and yellow rust.In line with the rust resistance breedingstrategy described previously,researchers from CIMMYT andAustralia are looking for minor genesto create durable resistance. ForCIMMYT’s partners in theinternational wheat improvementsystem, the value of this project is clear.For Australia, this work will provevaluable in the event that rustresistance in its wheat varieties (largelybased on major genes) breaks down, ashas occurred on occasions in the past.

The second project in the CRCcollaboration focuses on introducing,via transgenics, resistances to somefungal pathogens of wheat and thencharacterizing their effects. Animportant aspect of this work is toincrease transformation efficiencies,which were low at the outset. Rates oftransformation have been significantlyincreased (efficiency was 0.2-0.4%before; now it averages about 1% andmay reach 5% in the near future), andresearchers are proceeding with theother objectives.

By collaborating with the manyinstitutes involved with the CRC thatare leaders in molecular genetics inwheat, CIMMYT can tap into theirexpertise in ways that will greatly

benefit many of our partners in theinternational wheat improvementsystem. Australia will also see positiveresults from the collaboration.According to the last annual report ofthe CRC for Molecular Plant Breeding,“CIMMYT’s global field programprovides CRC scientists with theopportunity to evaluate germplasmand populations in a wide range ofenvironments. This makes it mucheasier for researchers to developmolecular approaches to the isolationof traits than if they were limited solelyto Australia’s agro-ecologicalenvironments” (CRC for MolecularPlant Breeding 1998).

Crop and Natural

Resource

Management

Research

When combined with robust, highlyproductive crop varieties, it is notuncommon for improved managementpractices to raise farmers’ yields twiceand even three times. Strategic researchon crop and natural resourcemanagement leads to improvedfarming practices and more sustainablemaize and wheat production systems.Such research involves a complexiteration of field studies, crop and soilmodeling, the use of geographicinformation systems, and remotesensing. At CIMMYT, agronomists areexamining nutrient auditing andstrategic fertilizer use; appropriatestrategies for replenishing soil organicmatter (such as green manures andcrop residues); the development of

19

suitable crop rotations; reducedtillage; and integrated pest and weedmanagement. Some of these strategiesare described in the sections thatfollow.

Improved InputUse EfficiencyCombining input efficiency with highyield potential in new cultivars willallow a farmer to benefit from thesecultivars over a wide range of inputlevels. Selection for yield potentialunder medium to high levels ofnitrogen has indirectly increased theefficiency of N uptake in CIMMYTwheats. Recently released CIMMYTbread wheat cultivars require less N toproduce a unit of grain than cultivarsreleased in previous decades (Ortiz-Monasterio et al. 1997). The increase innitrogen use efficiency is shown inFigure 2. Under low N levels in thesoil, N use efficiency increased mainlydue to a higher N uptake efficiency—the ability of plants to absorb N fromthe soil—whereas under high Nlevels, the N utilization efficiency—the capacity of plants to convertabsorbed N into grain yield—increased.

A study initiated in 1994 evaluatedchanges in soil nutrients and gasemission before and after fertilizerapplications and compared alternativeways of applying nitrogen (Matson,Naylor, and Ortiz-Monasterio 1998).The experiment compared thecommon practice of Yaqui Valleyfarmers with alternatives thatincluded reducing the amount ofnitrogen applied and changing thetiming of its application. Theresearchers found that with thefarmers’ practice, relatively high levelsof nitrogen were lost into the

atmosphere when nitrogen came intocontact with irrigation water, evenbefore the wheat crop was in theground. The best practice reduced theamount of nitrogen (from 250 to 180kg/ha, one-third applied at plantingand two-thirds six weeks later) andproduced yields and grain qualitysimilar to those obtained under thefarmers’ practice. The best alternativepractice also saved US$ 55-76/ha(equivalent to saving 12-17% in after-tax profits). The study shows that it ispossible to reduce nitrogen gasemissions and fertilizer losses throughappropriate agronomic practices andat the same time maintain yields.

Bed Planting SystemsA reduced tillage system developed byfarmers and researchers in Mexico’sYaqui Valley is showing its potentialthere and in other irrigated wheatproduction environments. In thissystem, a crop is grown on raised bedsthat are divided by furrows forirrigation. No soil inversion tillage isused on the beds. Crop residues are

8

7

6

5

4

3

21950 55 60 65 70 75 80 85

Genotype year of release

Grain yield (t/ha)

Figure 2. Grain yield of the historical seriesof bread wheats at Cd. Obregón, Mexico, at0 and 300 kg/ha N application.

300 kg N

0 kg N

20

chopped and left on the surface of thebeds. The system has severaladvantages for farmers and theenvironment, including:

• Nitrogen can be applied when andwhere the wheat plants can use itmost efficiently. Yields improve, andnitrogen losses into the environmentare significantly reduced.

• Water conservation improves. Aswater for agriculture becomes morescarce in the years to come, waterconservation practices will becomemore important for farmers.Researchers in South Asia and Chinareport a 30% savings in water usefrom using bed planting andimproved weed control.

• Weeds can be controlled bycultivating between the beds—reducing costs and the need forherbicide.

• Residues are returned to the soilwithout burning, which is beneficialto the environment.

• The beds can be used cycle aftercycle. Farmers avoid the financial andenvironmental costs of makingrepeated passes with a conventionalplow during land preparation.

Prototype machinery for this bedplanting system has been designedand tested in Mexico and in Asia. Theprototypes are modifications ofstandard agricultural equipment andare expected to be affordable for poorfarmers. Mexican farmers reportedlysave 30% on their production costswhen they use the bed plantingsystem. Some 10,000 farmers arethought to use the system in Mexico,and the number of farmers who areusing bed planting is growing in SouthAsia and China as well. In fact, inparts of China some farmers find the

technology so valuable that in theabsence of equipment they form thebeds by hand.

Farmer ParticipatoryResearchOver the past few years, CIMMYT hassignificantly increased its investmentin farmer participatory research fornatural resource management (that is,in the development of productivity-enhancing, resource-conservingpractices for maize and wheat systems,with beneficial impacts on soils, water,and agroecosystem diversity). Farmerparticipatory research is a tool for apurpose: the development ofsustainable practices that improveresource quality while raising systemproductivity. CIMMYT is movingaggressively to mainstream the use ofthis tool for these important ends. Forexample, in irrigated areas in northernMexico, CIMMYT has longcollaborated with farmers in thedevelopment of the bed plantingsystems described earlier.

In Asia, CIMMYT works with theother members of the Rice-WheatConsortium for the Indo-GangeticPlains to foster farmer experimentationon reduced and zero tillage strategiesfor establishing wheat after rice.Farmer groups have assessedalternative tillage and sowingimplements and wheat establishmentstrategies, and they have beenencouraged to develop their owninnovations and adaptations.Minimum tillage practices arespreading in Bangladesh, and farmersin the western part of the Indo-Gangetic Plains are beginning to usezero tillage. Farmers report earliersowing, higher yields with lower levelsof inputs, and improved possibilities

21

for diversifying cropping patternsaway from a continuous rice-wheatrotation—with numerousagroecological benefits.

In Bolivia, we are collaboratingwith farmer groups to develop zerotillage/mulch systems suitable forsmallholders (2-5 ha) in the high inter-Andean valleys. These farmersproduce one crop of wheat each yearin monoculture or in rotation withpotatoes, faba beans, peas, and/orbarley. Research focuses on evaluationof straw cover to increase rainfall useefficiency. Results are extremelyencouraging: crop residue retentiongenerally increases yields and reducesrisk, two important objectives forBolivia’s small-scale, subsistencefarmers. Researchers also participatein a project to develop a small,animal-drawn, no-till seed drill forsowing cereals into surface residues,and results are very positive(CIMMYT 1999).

Information

Management Tools

for Sustainable

Systems

Researchers have always believed inthe value of sharing information morewidely, but the limitations ofinformation technology have notmade this easy. CIMMYT now offers awidening array of informationmanagement tools to researchers inmany disciplines.

For example, the InternationalWheat Information System (IWIS) is arelational database available on CDwhich gives each genotype a uniqueidentifier and provides extensivepedigree and performance data. TheGenetic Resources InformationPackage (GRIP), designed inconjunction with Australian partners,allows IWIS users to locate seedsamples in wheat germplasm stocksin a number of collections around theworld and provides an abbreviatedversion of the IWIS pedigrees. TheInternational Crop InformationSystem (ICIS) is a data managementtool that builds on IWIS. It containsinformation on several crops inaddition to wheat. The core of ICIS isa relational database structure thatstores data on plant genetic resources,pedigrees, field and laboratoryevaluations (including molecularinformation), and auxiliary data onlocations, institutions, and people.Simple geographic informationfunctions are being incorporated intoICIS, and a tool for exporting data tocrop simulation models is also underdevelopment.

One challenge to sharinginformation more widely is toprovide access to cutting-edgegeographic information system (GIS)tools for non-GIS users, especiallythose in Africa. African researchersneed spatially referenced data onclimate, soils, infrastructure, cropdistribution, and the natural resourcebase, in part to ascertain the extent towhich their site-specific research mayhave relevance to larger areas. TheAfrica Country Almanacs containsuch base data, along with the mostcommonly requested maps, plussearch and viewing tools, on a single

22

compact disc. Almanacs have beendeveloped for 12 African countries,5

some of which have requested follow-on demonstrations and training fortheir research staff. Now all researcherscan have access to these powerful GIStools, not just a few specialists in acentral office.

The Spatial Characterization Tool(SCT) developed by CIMMYT andTexas A & M University goes a longway towards addressing the problemof “site specificity” in natural resourcesmanagement research. Site researcherscan now quickly perform “sitesimilarity analysis,” identifying areaswith environments resembling that oftheir site. When applied to sites inBolivia, this analysis uncoveredenvironmentally similar areas withinBolivia; in neighboring countries (e.g.,Chile, Brazil); within the Americas(e.g., Mexico); and even in otherregions of the world (Ethiopia,Lesotho). Scientists in these diverselocations find that they have much toshare about technology performanceand the consequences of technicalchange for system productivity andsustainability.

These information managementtools help encourage researchintegration, explore the prospectiveperformance of new technologies, andovercome site specificity. However, likeall information management tools, theyneed data. A final challenge is how topreserve, organize, and make availableto researchers the rich array of dataoften generated by research,particularly in natural resourcemanagement research. CIMMYT is

developing an answer to this set ofchallenges: the Sustainable FarmingSystems Database (SFSD). Non-governmental organizations are usingthe SFSD prototype to organizeinformation on the global experiencewith green manure cover crops. As theSFSD matures, its uses will bevirtually infinite.

Conclusions

The strategies we have just outlinedcould make the difference between asustainable future, with food andeconomic opportunity available forthe majority, and a future of scarcity,with survival seriously compromisedfor most people. Successful,sustainable agriculture can help createthe purchasing power andemployment that will ensure foodsecurity and help eradicate poverty.We believe that the risks of ignoringagricultural development will be farhigher than the risks of deciding tocreate a sustainable future for us all.

The world has faced a similarchoice before, when a decision wasmade to sow the new semidwarfwheats in India in the hope that theirhigher yields would prevent a famineas great as the devastating Bengalfamine of 1943. That decisiontransformed agriculture and the waythat agricultural research wasconducted. Today CIMMYT and itspartners join forces in one of theworld’s most ambitious endeavors:we participate in a global wheatimprovement system that continues tobetter the lives of millions of poor5 Including three important wheat

producing nations: Ethiopia, Kenya,and Zimbabwe.

23

farmers and consumers indeveloping countries. The impact ofthat system is well documented(Byerlee and Moya 1993; Marediaand Byerlee 1999; CIMMYT 1999). Inthe most recent period, 1991-97,almost 90% of the spring breadwheat varieties released by nationalagricultural research systems hadCIMMYT ancestry (Figure 3).Virtually all (98%) of the springdurum wheats released by nationalprograms in 1991-97 had CIMMYTancestry (Figure 4). Farmers nowplant almost 80% of the developingworld’s spring bread wheat area toCIMMYT-related wheats (Figure 5).

CIMMYT crosses(some re-selected byNARSs), 56%

Figure 3. Ancestry of spring bread wheatvarieties released by national programs,1991-97.Source: CIMMYT wheat impacts database.

NARS crosses with atleast one CIMMYTparent, 28%

NARS crosseswith someknownCIMMYTancestry, 5%

NARSsemidwarfswith otherancestry, 8%

Tallvarieties, 3%

CIMMYTcrosses, 77%

Figure 4. Ancestry of springdurum wheat varietiesreleased by nationalprograms, 1991-97.Source: CIMMYT wheatimpacts database.

NARS crosses withat least one CIMMYTparent, 19% NARS crosses

with someknownCIMMYTancestry, 5%

Tallvarieties, 2%

100

80

60

40

20

0

ChinaIndia

Other Asia

WANA

All Asia

Sub-Saharan Afri

ca

Latin A

merica

Developing countri

es

Figure 5. Area planted to spring breadwheat in developing countries, 1997.Source: CIMMYT wheat impactsdatabase.

Percentage of total springbread wheat area

Unknown

Landraces

Tall with pedigree

Other semidwarf

Any CIMMYTancestor

At least oneCIMMYT parent

CIMMYTcross

24

A New Research Paradigmfor New Research ImpactsThese research impacts are reassuring,but much remains to be done. Whenour colleague Norman Borlaugaccepted the Nobel Peace Prize for hisachievements in bringing about theGreen Revolution in wheat, hecautioned that the Green Revolution“has not transformed the world intoUtopia. None are more keenly aware ofits limitations than those who started itand fought for its success. . . . Aboveall, I cannot emphasize too strongly thefact that further progress depends onintelligent, integrated, and persistenteffort” (CIMMYT 1970).

Borlaug’s observation remains true.If we are to make progress towardsustainable food security, we must takehis advice and change the way weplan, conduct, and communicate aboutresearch. We must blend veryspecialized research disciplines inteams of scientists seeking appropriateoutcomes that have an immediateimpact in farmers’ fields. It is fromthese fields that food supplies mustcome for the foreseeable future. Thefarmer is the ultimate systems-orientedoperator, juggling biological, economic,environmental, and social factors. Insuch circumstances, isolatedinterventions are of limited value atbest; all too often, they make thingsworse.

These interventions will be basedon a new, integrative researchparadigm that focuses on the elementsof the GXEXMXP equation mentionedearlier: the best genotypes (G), in theright environments (E), underappropriate crop management (M),generating appropriate outcomes forpeople (P). Everyone who seeks to

foster sustainable agriculture indeveloping countries should recognizethe interdependence of these factors,because most organizations bythemselves cannot contribute fully toeach aspect of GXEXMXP. Partnershipsand consortia that assemble the bestpossible teams to execute the GXEXMXPapproach will underpin the timely andsuccessful achievement of sustainablefarming systems and future foodsecurity.

The Shape of Thingsto ComeGiven these requirements, what willagricultural research look like in thenew millenium? Every member of theinternational wheat improvementsystem—and the farmers andconsumers who depend on it—will beaffected by changes in internationalresearch in the years to come. Whichforces are likely to shape the way thatresearch is done—either bycontributing to or detracting from theintegrative research paradigm we havejust described?

For decades, collaboration has beenthe mainspring of the internationalwheat improvement system. None ofthe achievements described in thispaper could have been attainedwithout it. Gains from conventionalbreeding will continue to be significantin the next two decades or more(Duvick 1996), but these are likely tocome at a higher cost than in the past.Research managers and policy makersare increasingly concerned that thevery open, collaborative networks thathave sustained the wheat improvementsystem will become far morecircumscribed in coming years.

25

Rasmussen (1996) has stated thatnearly half of the progress made bybreeders in the past can be attributed togermplasm exchange. In recent surveysof wheat breeders (Braun et al. 1998;Rejesus, van Ginkel, and Smale 1996),more than 80% of respondentsexpressed concern that plant varietyprotection (PVP) and plant or genepatents will restrict access togermplasm, with deleteriousconsequences for future breedingachievements. Regional andinternational nurseries are an efficient,low-cost means of gathering data fromvaried environments and exposinggermplasm to diverse pathogenselection pressures, while providingaccess to germplasm and promotinggermplasm exchanges. Breeders usecooperative nurseries extensively intheir crossing programs, but thenumber of such nurseries has beengreatly reduced during the pastdecade, partly because of increasingrestrictions on germplasm exchange.

Recent developments inbiotechnology for plant improvementhave motivated much of the concernover PVP and other forms ofintellectual property rights (IPR), aswell as concern over germplasmexchange and developing nations’access to novel agriculturaltechnologies. That debate promises topale in comparison to anotherbiotechnology-inspired debate,however, that has been prominent inthe media.

The debate over the ethical uses ofbiotechnology has shifted to theagricultural sector. A furor overgenetically modified plants (focusingon uncertainty over their potentialeffects on human health and the

environment) has swept acrossEurope, where “the public’sperception of risk far outweighs itsview of the possible benefits” (TheEconomist, 19 June, 1999). Withindevelopment circles, some argue that itis too risky to use genetic engineeringto solve poor people’s problemsbecause we may be unaware of futureside effects. Others question whether itis ethical to withhold solutions toproblems that cause millions ofchildren to die from hunger andmalnutrition. Clearly we must seekacceptable levels of biosafety beforereleasing products from modernscience, but it is critical that the risksassociated with the solutions beweighed against the ethics of notmaking every effort to solve food andnutrition problems.

These highly public—and highlycharged—debates make it easy to losesight of another trend in the researchenvironment that is almost moreworrying. For a host of reasons, manynational agricultural research systemshave become weaker over the past twodecades rather than stronger. At theinternational level, public support forbroad research initiatives, such asCIMMYT’s improvement of wheatgermplasm for the majorenvironments in the developingworld, has diminished as publicresearch investments have increasinglyfocused on more narrowly targetedprojects. Under these circumstances,can we reasonably expect the publicsector to be an effective advocate onbehalf of the poorest constituents ofsociety? Will the declining resourcescommanded by the public sectorinterfere with germplasm testing and

26

exchange even more than the trendsdescribed earlier? Given the vastresources commanded by privateresearch organizations, what is thefuture role of the public sector in cropimprovement research?

Despite these uncertainties in theresearch environment, our ultimateobjective remains clear. We know thatto ensure food security in the 21st

century, the sustainable intensificationof agriculture in farmers’ fields isessential. With 200 people added eachminute to our population, and with allof us, rich and poor alike, dependenton a shrinking agricultural resourcebase, sustainable intensification is theonly practical and appropriate choicefor the foreseeable future. The new

millenium holds out incrediblepromise—superior technology,unprecedented access to information,economic growth—but if these serveonly to widen the gap between the“haves” and “have-nots,” between theNorth and South, then what will wehave gained? Of the many issuessurrounding the future of internationalagriculture, this is perhaps the mostimportant. It is the central issue thatmotivates CIMMYT’s research agenda,and it will remain at the forefront of allof our future efforts.

27

References

Borlaug, N.E., and C.R. Dowswell. 1997. The acidlands: One of agriculture’s last frontiers. InA.C. Moniz et al. (eds.), Plant-Soil Interactionsat Low pH. Brazil: Brazilian Soil ScienceSociety. Pp. 5-15.

Bramel-Cox, P.J., T. Barker, F. Zavala-Garcia, 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: CSSA and ASA. Pp. 29-56.

Braun, H.-J., H. Ekiz, V. Eser, M. Keser, H. Ketata,G. Marcucci, A.I. Morgounov, and N.Zencirci. 1998. Breeding priorities of winterwheat programs. In H.-J. Braun, F. Altay,W.E. Kronstad, S.P.S. Beniwal, and A. McNab(eds.), Wheat: Prospects for Global Improvement.Dordrecht, the Netherlands: KluwerAcademic Publishers. Pp. 553-560.

Byerlee, D., and P. Moya. 1993. Impacts ofInternational Wheat Breeding Research in theDeveloping World, 1966-1990. Mexico, D.F.:CIMMYT.

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.

Ciaffi, M., L. Dominici, D. Lafiandra, and E.Porceddu. 1992. Seed storage proteins ofwild wheat progenitors and theirrelationships with technological properties.Hereditas 116: 315-322.

CIMMYT. 1970. The Green Revolution: Peace andHumanity. Norman E. Borlaug, 1970 NobelPeace Price. Mexico, D.F.: CIMMYT.

CIMMYT. 1986. Veery ‘S’: Bread Wheats for ManyEnvironments. Mexico, D.F.: CIMMYT.

CIMMYT. 1999. A Sampling of CIMMYT Impacts,1999: New Global and Regional Studies.Mexico, D.F.: CIMMYT.

Cooper, M., D.E. Byth, and D.R. Woodruff. 1994.An investigation of the grain yieldadaptation of CIMMYT wheat lines to waterstress environments in Queensland. II.Classification analysis. Agricultural Research45: 985-1002.

CRC (Cooperative Research Centre) forMolecular Plant Breeding. 1998. CooperativeResearch Centre for Molecular Plant Breeding:Annual Report 1997/98. Adelaide, Australia:CRC for Molecular Plant Breeding.

Devos, K.M., and M.D. Gale. 1997. Comparativegenetics in the grasses. Plant MolecularBiology 35: 3-15.

Duvick, D.N. 1990. Ideotype evolution of hybridmaize in the USA, 1930-1990. In Vol. II ofATTI Proceedings, National Maize Conference:Research, Economy, Environment. Bologna,Italy: Centro Regionale per laSperimentazione Agraria, Pozzuolo delFriuli Edagrocole s.p.a. Pp. 557-570.

Duvick, D.N. 1992. Genetic contributions toadvances in yield of US maize. Maydica 37:69-79.

Duvick, D.N. 1996. Plant breeding, anevolutionary concept. Crop Science 36: 539-548.

Economist. 1999. “Genetically Modified Food:Food for Thought,” 19 June.

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.

Fischer, R.A., D. Rees, K.D. Sayre, Z. Lu, A.G.Condon, A. Larque-Saavedra, and E. Zeriger.1998. Wheat yield progress is associatedwith higher stomatal conductance, higherphotosynthetic rate, and cooler canopies.Crop Science 38: 1467-1475.

ICARDA (International Center for AgriculturalResearch in the Dry Areas). 1993. CerealProgram. Annual Report for 1992. Aleppo,Syria: ICARDA.

Lafiandra, D., M. Ciaffi, and S. Benedetelli. 1993.Seed storage proteins of wild wheatprogenitors. In A.B. Damiana (ed.),Biodiversity and Wheat Improvement. NewYork: John Wiley. Pp. 329-340.

Lagudah, E.S., and G.M. Halloran. 1988.Phylogenetic relationships of Triticumtauschii the D genome donor to hexaploidwheat. I. Variations in HMW subunits ofglutenin and gliadins. Theoretical and AppliedGenetics 75: 592-598.

Maredia, M.K., and D. Byerlee (eds.) 1999.CIMMYT Research Report No. 5. Mexico, D.F.:CIMMYT (forthcoming).

Matson, P.A., R. Naylor, and I. Ortiz-Monasterio.1998. Integration of environmental,agronomic, and economic aspects offertilizer management. Science 280: 112-114.

McFadden, E.S. 1930. A successful transfer ofemmer characters to vulgare wheat. Journal ofthe American Society of Agronomy 22: 1020-1034.

Ortiz-Monasterio, J.I., K.D. Sayre, S. Rajaram,and M. McMahon. 1997. Genetic progress inwheat yield and nitrogen use efficiencyunder four nitrogen rates. Crop Science 37:898-904.

28

Payne, P.I., K.G. Corfield, L.M. Holt, and J.A.Blackman. 1981. Correlations of certain high-molecular-weight subunits of glutenin andbreadmaking quality in progenies of sixcross of bread wheat. Journal of the Science ofFood and Agriculture 32: 51-60.

Pinstrup-Andersen, P., and R. Pandya-Lorch.1997. Can Everybody Be Well Fed by 2020without Damaging Natural Resources? FirstDistinguished Economist Lecture. Mexico,D.F.: CIMMYT.

Pogna, N.E., J.C. Autran, F. Mellini, D. Lafiandra,and P. Feillet. 1990. Chromosome 1B-encoded gliadins and glutenin subunits indurum wheat: Genetics and relationship togluten strength. Journal of Cereal Science 11:15-34.

Rajaram, S., and M. van Ginkel. 1995. Wheatbreeding methodology: Internationalperspectives. Paper presented at the 20th

Hard Red Winter Wheat Workers Workshop,Oklahoma City, USA.

Rasmussen, D.C. 1996. Germplasm is paramount.In M.P. Reynolds, R. Rajaram, and A. McNab(eds.), Increasing Yield Potential in Wheat:Breaking the Barriers. Mexico, D.F.: CIMMYT.Pp. 28-35.

Rees, D., K. Sayre, E. Acevedo, T.N. Sanchez, Z.Lu, E. Zeiger, and L. Limon. 1993. CanopyTemperatures of Wheat: Relationship with Yieldand Potential as a Technique for EarlyGeneration Selection. Wheat Special ReportNo. 10. Mexico, D.F.: CIMMYT.

Reeves, T.G. 1998. Sustainable Intensification ofAgriculture. Mexico, D.F.: CIMMYT.

Reeves, T.G. 1999. Intensification for the ninebillion. Nature Biotechnology (supplement) 17:BV33-BV34.

Reeves, T.G., P. Pinstrup-Anderson, and R.Pandya-Lorch. 1997. Food Security and theRole of Agricultural Research. Mexico, D.F.:CIMMYT.

Rejesus, R.M., M. van Ginkel, and M. Smale.1996. Wheat Breeders’ Perspectives on GeneticDiversity and Germplasm Use. Wheat SpecialReport No. 40. Mexico, D.F.: CIMMYT.

Reynolds, M.P., M. Balota, M.I.B. Delgado, I.Amani, and R.A. Fischer. 1994. Physiologicaland morphological traits associated withspring wheat yield under hot, irrigatedconditions. Australian Journal of PlantPhysiology 21: 717-730.

Reynolds, M.P., R.P. Singh, A. Ibrahim, O.A.A.Ageeb, A. Larque Saavedra, and J.S. Quick.1998. Evaluating physiological traits tocomplement empirical selection for wheat inwarm environments. Euphytica 100: 85-94.

Rosegrant, M.W., M.A. Sombilla, R.V. Gerpacio,and C. Ringler. 1997. Global food marketsand US exports in the twenty-first century.Paper presented at the Illinois World Foodand Sustainable Agriculture ProgramConference, Meeting the Demand for Foodin the 21st Century: Challenges andOpportunities, 28 May, University ofIllinois, Urbana-Champaign.

Smale, M., R.P. Singh, K. Sayre, P. Pingali, S.Rajaram, and H.J. Dubin. 1998. Estimatingthe economic impact of breedingnonspecific resistance to leaf rust in modernbread wheats. Plant Disease 82(9): 1055-1061.

Snape, J.W. 1998. Golden calves or whiteelephants? Biotechnology for wheatimprovement. Euphytica 100: 207-217.

Uddin, N., B.F. Carver, and A.C. Clutter. 1992.Genetic analysis and selection for wheatyield in drought-stressed and irrigatedenvironments. Euphytica 62: 89-96.

UNDP (United Nations DevelopmentProgramme). 1997. Human DevelopmentReport 1997. [www.undp.org/hdro/e97over.htm].

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 ofwheat in early, late, or continuous droughtconditions. Euphytica 100: 109-21.

Villareal, R.L., E. del Toro, A. Mujeeb-Kazi, andS. Rajaram. 1995. The 1BL/1RS chromosometranslocation effect on yield characterizationin a Triticum aestivum L. cross. Plant Breeding114: 497-500.

Waines, J.G., and P.I. Payne. 1987.Electrophoretic analysis of the high-molecular-weight glutenin subunits ofTriticum monococcum, T. uartu, and the Agenome of bread wheat (T. aestivum).Theoretical and Applied Genetics 74: 71-76.

William, M.D.H.M., R.J. Peña, and A. Mujeeb-Kazi. 1993. Seed protein and isozymevariations in Triticum tauschii (Aegilopssquarrosa). Theoretical and Applied Genetics 87:257-263.

World Bank. 1997. World Development Indicators1997. New York: Oxford University Press.

Zavala-Garcia, F., P.J. Bramel-Cox, J.D. Eastin,M.D. Witt, and D.J. Andrews. 1992.Increasing the efficiency of crop selectionfor unpredictable environments. CropScience 32: 51-57.

Maarten van Ginkel, head of bread wheat

breeding at CIMMYT, holds one of the

large-spiked wheats (right) that promise

to raise yields in wheats. On the left he

holds a normal wheat spike. (See page 7.)