DeVries &Toenniessen_ Securing the Harvest.pdf

Acknowledgements

This book grew out of a 2-year exploration conducted by the Food Security theme ofThe Rockefeller Foundation focusing on the potential for crop genetic improvement tocontribute to food security among rural populations in Africa. The exploration carriedthe authors to ten countries of sub-Saharan Africa and a number of related national,regional, and international meetings on several continents. Along the way, innumerableindividuals farmers, researchers, seed merchants, policy experts and others contrib-uted their views, comments and experiences related to crop improvement in Africanagriculture, and the authors are very grateful for their assistance.

In particular, we would like to thank Foundation colleagues Gordon Conway,Bob Herdt, John Lynam, John OToole and Ruben Puentes for reading through themanuscript and sharing their views. David Jewell and Gebisa Ejeta also read early draftsand gave useful comments. In addition, we would like to express our gratitude to thefollowing individuals who assisted with the study by sharing their views and providinginformation: Marianne Banziger, Jeffrey Bennetzen, Malcolm Blackie, RonnieCoffman, Joel Cohen, Ken Dashiell, Alfred Dixon, Peter Ewell, John Hartmann, TomHash, Dave Hoisington, Lee House, Justice Imanyowa, Jane Ininda, Saleem Ismael,Monty Jones, Richard Jones, Bill Kiezzer, Laurie Kitch, Jenny Kling, Dennis Kyetere,Isaac Minde, Larry Murdoch, Patricia Ngwira, Hannington Obiero, Joseph Ochieng,Moses Onim, James Otieno, Yvonne Pinto, Kevin Pixley, Fred Rattunde, DarrellRosenow, B.B. Singh, B.N. Singh, Elizabeth Sibale, Ida Sithole, Margaret Smith,Aboubacar Toure, Lamine Traore, Wilberforce Tushemereirwe, Eva Weltzien and JohnWhyte. Finally, we would like to express our sincere appreciation to Sarah Dioguardiand Mulemia Maina, who provided excellent care and technical assistance in preparingthe manuscript.

Inevitably, when attempting to address as broad a range of issues as biotechnologyto seed production in a number of important crops, mistakes and discrepancies willoccur, both in terms of the facts gathered and the assertions made. Although the authorshave tried to avoid these, they apologize in advance for those that remain, and take fullresponsibility for them.

ix

A4138:AMA:DeVries:First Revise:15-Oct-01 Prelims9

Executive Summary

Food crops grown under low-input, rain-fed conditions in sub-Saharan Africa areaffected by a wide range of biological and environmental constraints, but remain thebest, if not the only, means of improving food security among the rural poor. To reachmaturity and yield well, crop varieties must be able to resist or tolerate these stressfactors. Due to wide variation in environmental conditions over space and time, theparticular set of constraints which operate in any given area is continually changing.Moreover, local processing and consumption needs often exert additional qualityrequirements in order for improved crop varieties to be adopted by small-scale farmers.To be successful, breeding programmes for Africa must take into consideration thisvariation and relevant varietal preferences of farmers. By analysing these requirements,selecting appropriate parental materials, and making selections under relevant localconditions with regular farmer input, new varieties with the right combination ofgenetic resistances and tolerances can be produced.

This kind of approach differs significantly from the methodology of selecting forhigh yield potential and broad adaptation which continues to give good results in morestable and more highly modified agricultural environments such as those in developedcountries and the irrigated regions of developing countries. The major implication is aneed for more localized, agro-ecology-based breeding programmes, where the principalobjective is to assemble a set of traits that reduce yield losses and thereby confer greateryield stability. Over time, yield-enhancing genes may still be added and make a signif-icant contribution to overall performance, but the emphasis during the current phase ofbreeding programmes should be placed on critical resistance factors.

The need to develop a range of improved varieties for Africa, each well adapted tolocal conditions, argues strongly for giving priority to well-funded and staffed cropbreeding programmes at the national level. Country-level programmes have lower costsand are able to deploy larger numbers of teams which can operate in close proximity tothe various agro-ecologies that need to be covered by any given programme.

xiii


International agricultural research centres (IARCs) have a major role to play infacilitating the development of fully capable national agricultural research systems(NARSs) able to produce the steady flow of new offerings required by farmers. Inaddition, international centres and advanced research institutes should devotesignificant resources and attention to the more difficult, intractable constraints of cropproduction which affect the important crop species of Africa. Such intractableconstraints are numerous and have not been solved despite considerable effort usingconventional techniques. By combining their talent and resources, and drawing on thestrengths of biotechnology, international centres and advanced research institutes maynow be able to overcome many, if not most, of these difficult problems.

Biotechnology remains a highly underdeveloped resource for improved foodproduction in Africa, largely due to underinvestment by governments and internationaldonors. Africa already has a number of scientists trained in biotechnology who areunable to utilize their knowledge owing to lack of facilities and operating funds. Sincethis situation may continue for some time to come, development of fully functionalbiotechnology capacity in all NARSs is not likely. However, those countries that canadequately staff both conventional and molecular breeding units should be encouragedto do so. Tissue culture of clonally propagated crops has already proved its valueto agriculture in Africa. A second application of biotechnology which could provecost-effective in the short to medium term is marker-aided selection for a range of traits,with the primary objective being to combine as many resistance traits as are required tomaximize crop performance under low-input conditions. Finally, as national biosafetyregulations and systems become operational, it will become more logical to invest innational expertise and facilities for crop genetic engineering, whereby critical resistancetraits may be transferred directly into otherwise well-adapted varieties.

Localized, agro-ecology-based crop improvement schemes need to be supportedby similarly oriented seed enterprises. In Africa, investment in the seed sector hashistorically been very low, in part influenced by the poor success record of large seedcompanies on the continent. Large, monopolistic seed companies have perceived littleadvantage in pursuing the locally directed breeding programmes needed to developa range of varieties adapted to the various niches created by environmental variation.Multinational seed companies that rely solely on their own offshore breeders and genebanks find it difficult to overcome the adaptation barriers of Africa; and their historicalreluctance to commercialize germplasm under licensing agreements with the publicsector further diminishes the attraction of operating in Africa.

Conversely, smaller entities operating in a competitive environment that allythemselves to NARS breeding programmes for access to new varieties may perform wellwith respect to small farmers interests. Their most obvious limitations size and lack ofcapital can serve as an effective entry point for governments, private investors anddonor agencies. Limited production of foundation seed is one bottleneck to the growthof this and related models for development of the seed industry. More harmoniousregulatory structures across the region are also needed.

Taken together, the combination of new science, new ways of working withfarmers, new opportunities for private sector seed supply, and a greater appreciation ofAfricas diverse agro-ecologies represent a new era in crop genetic improvement forAfrica. Old arguments for products already being on the shelf lose their meaning in

xiv Executive Summary


view of what scientists and farmers can achieve today, if the needed effort and resourcesare put forward.

In this context, the importance of responsive, relevant public policy in the further-ance of a healthy, functional germplasm sector in Africa can hardly be overestimated.Public policy makers need to be committed to solving the problem of food insecurity onthe continent, and to employing relevant, up-to-date policy that can strengthen thebreeding and seed sectors. The worldwide biotechnology debate has provided the latestopportunity to put African agriculture in the spotlight and emphasize the need to movepolicy structures forward rapidly and responsibly. A major tenet of these changes mustbe to encourage private investment of all kinds in the seed sector. Another is thereinforcement of public sector capacity in crop improvement, using both conventionaland molecular techniques. The establishment of an effective set of biosafety regulationsis also critical to taking advantage of recent advancements in crop improvement.

In spite of its potential, genetic improvement of crops will always face limitationswith regard to what it can offer to farmers in regards to their levels of productivity. Nomatter what efficiencies genetic enhancement is able to build into crop plants, they willalways draw their nutrition from external sources, and this places enormous importanceon the investments that can be made in the soils of Africa. Overall improvements in agri-cultural productivity are likely to move in tandem with improvements made in the man-agement of soil nutrients by African farmers. Shortfalls in the level of nutrient supplythat are possible through the uses of organic methods must be complemented by makingfertilizer broadly more accessible to small-scale farmers. Because of the need to demon-strate the potential of the combined effects of genetic improvement and improved soilfertility, crop improvement initiatives and soil fertility management programmes shouldoperate in similar environments and test their results on the same or similar sites.

These policy and technology innovations can combine well with the revolution infarmer participation in agricultural research. One very critical entry point for farmerparticipation is the need to understand better the various agro-ecologies that can betargeted by public breeding programmes. Farmers are the best source of informationregarding the number and prioritization of production constraints, as well as the spatialdistribution of differing agro-ecologies. In view of the importance and complexity oftheir preferences for processing, taste, growth habit, and multiple uses of crop plants,farmers also need to be made part of the process of varietal selection. While there is noset procedure for farmer participation in breeding schemes in Africa, it seems obviousthat breeding programmes which operate in close proximity to farmers and their base ofknowledge will have definite advantages over those which do not involve farmers.

Within this rapidly evolving professional context, crop genetic improvement can beviewed as a highly underexploited resource for improving food security among Africasmajority, rural populations. Indeed, with late-maturing, low-yielding crop varietiesdominating the farming systems of much of Africa, crop genetic improvement stillhas the potential to play an important role in the development of more productive agri-cultural systems throughout the continent.

A new paradigm for germplasm improvement in Africa, and indeed in other regionsof the developing world, can be envisioned. It is a paradigm driven first and foremost bythe urgent need for food security in Africa among a growing population of very poor,rural people who have been left behind by globalization and the interests of the private

Executive Summary xv


sector. The impetus for understanding the details of their needs in terms of better, moreresilient crops leads directly to the application of an enriched set of technologies forcrop improvement, including conventional, field-based selection and laboratory-basedmodification and enhancement of the germplasm. Public sector technology develop-ment in this paradigm is linked directly to a broad interface of private initiative throughnon-governmental organizations (NGOs), farmers associations and private business.And, it is backed up by the commitment to serving the peoples needs through seeddistribution by an efficient seed sector.

Providing the crop varieties needed to improve food security across the vast conti-nent of Africa is an enormous challenge. No donor or national government acting alonewould be able to mobilize the commitment and resources necessary to make a majorchange in this area. Novertheless, there is very little likelihood that Africa will be foodsecure without an intensive, long-term programme of investment in crop improvementwhich takes advantage of the full range of approaches now available.

While it would be going too far to declare that improved food security throughhigher-performing crops throughout Africa is readily do-able, it is at least possible tobreak down the process into conceivable steps as follows:

1. Constructing the breeding teams within NARSs supported by IARCs.2. Delineating and classifying the agro-ecologies which merit targeting.3. Determining farmer preferences for new varieties.4. Employing appropriate parental materials and breeding methods aimed to producenew varieties within an acceptable time frame.5. Getting seed to farmers via public and private means.

Running concurrently with the process above, biotechnology studies aimed atdeveloping solutions to intractable problems can be initiated at any time. Products ofthose studies feed into step 4. In view of the recent, negative trend in food availabilityand child nutrition in sub-Saharan Africa, food security in Africa is one of the mostcritical challenges facing humankind today. The record of private investment during thepost-structural adjustment era does not present a convincing argument that growth inthe private sector alone will lead Africa out of poverty and food insecurity. The publicsector still plays an enormously important role in offering hope to the poor and excludedthroughout the continent. And within this grouping, public sector capacity in thegenetic improvement of food crops presents an exciting opportunity to make realprogress.

xvi Executive Summary


Foreword

One of the many outcomes of the global media debate on biotechnology has been aheightened level of awareness, and one wants to believe interest in how the food cropsthat provide our nutrition are developed, grown, and eventually end up on our tables.This is a positive outcome for several reasons. First, because agriculture in the developedworld, although playing a huge role in the way we live, tends to remain out of sight andout of mind for nearly all of us except the 2 or 3% who are farmers and farm workers.Second, because it has reminded us that food security is never a resolved issue. One wayor another, we have to keep on producing enough food for 6 billion people today and 8billion by 2025, or there could be mass starvation. Finally, the debate on biotechnologyhas provided spokespersons of the agricultural community with the opportunity ofexplaining to the rest of the world just how dependent we already are, with or withoutbiotechnology, on genetic improvement of food crops and on inputs such as fertilizer.

As the one remaining major world region where agriculture has yet to be trans-formed from subsistence, low-yield systems dependent on shifting cultivation toefficient, modern systems capable of producing regular surpluses, the question of cropimprovement is especially important to Africa. Africa is also the sole world region wheremany indices of food security have shown a serious decline in recent years. In the contextof continued high population growth and an increased emphasis on keeping Africasunique natural environment intact, it is clear that crop yields must be substantially andsustainably increased. As they have in all other parts of the world, more efficient,better-performing crop varieties can play an important role in achieving this goal.

This book came about as part of a major restructuring of The Rockefeller Founda-tion which has resulted in a renewed commitment to the poor and excluded of theworld, who have largely been left behind by globalization and economic growth. Thestudys Africa focus reflects a greater emphasis being placed by our Food SecurityProgramme on that part of the world bypassed by the Green Revolution. Its attentionto issues ranging from frontier research in biotechnology to participatory methods of

xi


seed dissemination via farmers groups reflects a greater concern with the application ofscience to the needs of the poor that result in real, positive changes in their lives andlivelihoods.

The title of Part I of this book, Biotechnology, Breeding, and Seed Systems for AfricanCrops: Re-thinking a 10,000-year-old Challenge, reflects what is an ambitious attempt bythe authors to encapsulate in a brief format our current understanding of the nature ofthe task of extending better-performing crop varieties to Africas farmers. While it isclear that any one group can only focus on selected portions of this process, it is hopedthat the opportunities identified can mobilize additional resources and generate newpartnerships which cover the full scope of the challenge ahead.

It has been of particular interest to me to note the important roles the authors fore-see for gaining a greater understanding of agro-ecologies in Africa and for the applicationof participatory methods as well as biotechnology. By generating new crop varieties withgreater yield stability, greater productivity and greater local acceptability, and by gettingthe new genetic resources into farmers hands through more responsive seed systems,they believe increased food security can be attained.

Gordon Conway

New YorkJanuary 2001

xii Foreword


I Biotechnology, Breeding,and Seed Systems forAfrican Crops: Re-thinkinga 10,000-year-old Challenge

A4138:AMA:DeVries:First Revise:15-Oct-01 Chapter I1

1 Introduction and Summary

Introduction

The final decades of the 20th century achieved the fastest growth of the global economyever recorded over a similar period. Rapid scientific and technological innovation,coupled with the opening up of economies throughout the world, permitted morepeople to improve their well-being than had ever been possible before over such arelatively brief period of time. Globalization, as the phenomenon of capital mobilityand global distribution of technology came to be called, brought employment andopportunity to innumerable groups of people who only one generation previously wereunable to imagine such changes.

For many of the worlds poor, the most immediate effect of global economicgrowth meant simply eating better and enjoying the many subtle but influential benefitsof adequate nutrition and improved health. World food prices fell dramatically, whilelife expectancy rose sharply. As living standards rose, greater numbers of children werealso able to attend school.

At the centurys close, however, it became apparent that not all the worldspopulation was equally swept along in the positive trend. A significant portion ofthe worlds population had failed to benefit from globalization. Furthermore, thewidespread rolling back of social services within the public sector, coupled with thewidespread belief that in the new world order everyones needs would be adequately metthrough the marketplace, meant that many members of this group had less chances ofescaping poverty than ever before.

The inability of a large segment of the worlds population to benefit from thecurrent socio-economic advances presents the world with intellectual and moralchallenges that cannot be ignored. Africa, it is widely accepted, lies at the core of thischallenge. While home to fewer total numbers of this left-behind group than Asia, italso embodies fewer of the factors necessary for inclusion in the growth-led process ofglobalization. In Africa, large portions of the population do not have access to sufficientfood. Economies are not growing at rates required to generate new opportunities for

3

A4138:AMA:DeVries:First Revise:15-Oct-01 Chapter 13

growing populations. The ensuing widespread frustration felt by local populations isproving a fertile ground for civil conflict and regional wars.

With upwards of 70% of Africas workforce engaged in farming, agriculturerepresents an important channel for extending new opportunities for improving thewell-being of hundreds of millions of people throughout the continent. Unfortunately,continued reliance on technologies and practices designed for a previous era means thatagriculture has largely become a trap for Africas rural population, guaranteeing a life ofpoverty and isolation.

It does not have to be this way. The arrival of the information age, combined withnew biological technologies, and new ways of linking people to them, means thatpeoples lives can be improved in a relatively brief period of time. This book explores theways to take advantage of the new capacity for global knowledge sharing and increasedpublic and private capital gains from the past decades, and direct them at one importantgroup of technologies crop genetic resources to broadly improve access to adequatefood and enhance the well-being of Africas rural poor.

This book is not intended as an analysis of all the activities ongoing in crop geneticimprovement in Africa. Rather, it is intended as a collection of observations obtainedfrom a wide range of geographical locations, as well as institutions involved in makingcrop genetic resources more valuable and more useful to African farmers. In makingthose observations, the authors also attempt to understand what has worked in Africa,what has not, and how the lessons learned might be grouped together to provide someguiding principles for crop genetic improvement (the term is used to imply all methodsavailable, including biotechnology, conventional breeding, and seed dissemination)work in the future. Nevertheless, they are the first to recognize that much needs to belearned in Africa, and that the views of many qualified people must yet be sought.

Improved food security, led by increased productivity among Africas many small-scale farmers, has been the aim of significant national and international effort in recentdecades. However, the relatively underdeveloped, non-globalized state of Africanagriculture presents scientists, farmers and development agents with challenges at anumber of levels. African agriculture at the close of the 20th century remains by andlarge an organic, living system, where biophysical signals within and between croppingsystems still pulse and exert checks and balances on the levels of success that can beenjoyed by any single organism within the system. Strategies for increasing cropproductivity which operate within this context must be significantly different from thoseapplied in modernized, highly manipulated agricultural systems of developed regions ofthe world. They must also differ substantially from strategies applied in Asia, previouslycited as a potential model for the challenge in Africa. This book attempts to show whythis is the case, and take a new look at the potential role of crop genetic improvement inmaking sustainable improvements in the food security status of poor rural people inAfrica.

Improved varieties of African crops are destined for cultivation in soils that are verylow in fertility and where attacks by pests and diseases and periodic drought oftenfurther reduce yields. These factors have led some to conclude that genetic improvementof African crops cannot result in major social benefits. Indeed, some popular argumentscontend that increasing food production can do little to stave off hunger.

But crop improvement within this context is not just about raising yieldthresholds, just as efforts aimed at making food more abundant in chronically food

4 Chapter 1


insecure regions may have little impact on national food balances. Increasing theamounts of food produced among poor farmers is aimed at improving nutritionand maximizing options among people whose options are few. Better crop varietiesfor African farmers also involve increasing yield stability and safeguarding themeagre investments of some of the worlds most vulnerable people. This book arguesthat real gains in food availability for the poor and excluded of Africa are possiblethrough publicly based, multi-tiered crop improvement strategies which are informedby farmers needs and attuned to the agro-ecologies in which the new varieties will beused.

The argument put forward in this text does not contend that better varieties aloneare the answer to food insecurity in Africa. Rather, more resilient, higher yieldingvarieties are viewed as an important component of a broadly improved and better-supported African farming environment. Improved varieties will inevitably performbetter on more fertile soils, just as farmers efforts at securing the harvest will go muchfurther within national and international policy environments that support and valuetheir livelihood.

Nevertheless, this study does recognize and seek to capitalize on the advantages ofseed and other planting materials in situations where, at present, little other assistancecan be offered to farmers. Seeds are animate technologies that can be easily transportedand transferred from one hand to another. Seed is also often the cheapest input available.Improved varieties, therefore, are frequently the only modernized input used by Africanfarmers. They are the first step in securing the harvest.

The frame of reference for this study is one which will be very familiar to many fieldresearchers and development agents working in Africa. It is that of a single mother ofseveral children whose primary means of income is a 1 ha plot of unimproved land on aneroded hillside. Depending on which part of the continent she is from, her principalcrop may be maize, sorghum, cassava, millet, rice, or banana. Inevitably, her farm willcontain other crops as well, such as cowpea, common bean, finger millet, groundnut,and if she is lucky, a few cash crops such as vegetables. From each harvest, she mustprovide for virtually all the needs of her family throughout the year, including clothing,health care, education costs and housing. Because she can afford few purchased inputs,the yield potential of her farm at the outset of the season is low she can expect toharvest a maximum of perhaps 2000 kg of produce. Meagre though it may be, in mostyears, through a wise combination of sales, barter and home consumption, she may beable to cope at this low level of productivity.

Figure 1.1ad depicts hypothetically her farms productivity potential under differ-ent levels of intervention with adapted, accessible technologies, including better cropvarieties and more fertile soils. During the course of any given season, innumerablethreats to the crops appear on the scene (Fig. 1.1a). In the case of maize, the threatsmight be drought, maize streak virus, stem borers, and the parasitic weed, Striga. If sherelies on cassava, the threats to her harvest may include African cassava mosaic virus,bacterial blight and green mites. Periodic drought during the season has a further,negative effect on yield. The impact of drought plus whatever combination of pests anddiseases attacks the crop in a given year can often reduce the average harvest on her farmby perhaps 5060%, to 1000 kg of harvestable produce. At this level of productivity, thefamily is on the edge of survival. If the losses are greater, or if disease enters the home,some members may not survive.

Introduction and Summary 5


6C

hapter1

A4138

Fig. 1.1. Strategies for securing the harvest in marginal farming zones of Africa.

6

The effect of crop varieties which resist and/or tolerate these constraints (Fig. 1.1b)can reduce such losses and raise her harvest above the theoretical survival line of1000 kg. Meanwhile, improved soil fertility, shown separately in Fig. 1.1c, could allowher to raise her initial potential harvest to perhaps 3000 kg, but without better varietiesthe harvest is still reduced to some point at or near the survival line by the end of theseason. Improved soil fertility and resilient crops combined (Fig. 1.1d), could provide herwith the kind of productive potential and yield stability necessary to raise her harvest toperhaps 2000 kg, a major improvement.

As uncomfortable as it may make us feel to contemplate the situation of this womanand her children, this is the reality of millions of farm families throughout Africa today.Certainly, they stand in need of development in the broadest sense. They need betterroads, better schools, better health care, and more employment opportunities. But theyalso need better crop varieties, and in particular varieties which are resilient to drought,low nutrient soils, insect pests and the myriad of diseases which attack crops in Africa.

Improving productivity securing the harvest in low-input systems wherefarmers cannot afford purchased inputs means delivering as many useful traits aspossible within the seed. The end product of these efforts, moreover, must be usable andacceptable by rural households. On a continent where upwards of 70% of the totalpopulation are engaged in farming, better and more resilient crops which produce alarger and more dependable harvest can be an effective strategy for delivering more foodand earning potential to those who need it most.

Directing science and technology at the ground-level needs of poor farmers maynot be the most effective way to increase food production on a national level. Better-offfarmers in more favourable farming environments may be quicker to adopt new tech-nologies and produce higher yields with them once they are in place. Nor are resilientcrop varieties a new idea. But unlike in Asia, rain-fed, marginal farming conditions arenot a secondary focus in Africa, to be targeted once the more favourable areas havebeen tapped. The difficult conditions and household scenarios like that faced by ourwoman on the hill and her children are the only target whose solution will bring aboutmeaningful change in the vast, rural areas of the continent. Recent observations of therevolution being brought about by globalization indicate they will not be deliveredunless very practical, results-oriented programmes are implemented by agriculturalresearch and development agencies within the public sector (Flavel, 1999; Persley andDoyle, 1999). It is an enormous challenge, made more difficult by the very limitedresources currently being put forward to address it. Understanding how the challengecan be approached putting together the pieces, if you will of some very promisingrecent advances in the science and methods of working with the poor would seem auseful subject to explore.

Summary

Recent years have seen vast improvements in our understanding of the genetic make-upof crop plants and the techniques available for enhancing them. It is now possible to domore for the woman on the hill than ever before. Indeed, the failure of the GreenRevolution to take root previously in Africa means that, in one form or another, most ofthis potential is yet to be realized. Nevertheless, accessing these advances and directing

Introduction and Summary 7


them toward the needs of the poor in an increasingly private sector-driven developmentagenda is a major challenge which requires support from many sides.

Crop genetic resources are assets that the poor and excluded can own andfurther modify to meet their needs. Experience and observation have shown that Africanfarmers are intensely interested in questions of crop variety performance. Most arealready engaged in informal variety trials of their own design. Their expertise can betapped in the search for resistance genes, in making selections, in growing out progenyand in adapting varieties to local conditions. Many of them can help deploy anddisseminate the new varieties which result.

African farmers are supported by a group of committed, well-trained scientists andtechnicians who understand well the tasks they face. Nevertheless, their numbers are asyet insufficient to ensure full success. Moreover, their lack of access to operating fundsregularly reduces the rate and extent to which their knowledge can be applied.Additional training is needed, especially in the newer techniques for genetic improve-ment and in understanding better Africas diverse agro-ecologies; and additionalfinancial support is required to allow them to put their strategies into action.

Some of the solutions are close at hand. For example, introgressing resistance tomaize streak virus in African maize populations should be a relatively simple task. Ifknown resistance genes were transferred into locally well adapted genotypes, maizeproduction across Africa might be increased by several million tonnes (see Plate 1).Achieving solutions to other constraints will require more complex, high-risk ventures.Downy mildew disease of millet is controlled by up to 17 genes (Hash et al., 1996), andthe fungal pathogen can rapidly evolve new pathotypes. Resistance in cereal crops to theparasitic weed, Striga, is so ephemeral a trait that researchers are working at transferringin more durable resistance genes from wild relatives (Ejeta et al., 2000; Kling et al.,2000).

This book attempts to consider both the broad context of the role of crop geneticimprovement in improving food security in Africa and the more specific, scientificchallenges inherent to improvement strategies within important crop species. As such, itis divided into two parts. This first part of the book looks at a range of human andenvironmental factors which condition efforts aimed at benefiting farmers throughimproved crop varieties, and then focuses on the discrete but interlinked roles of cropbreeding, biotechnology, and seed systems in developing and delivering new products tofarmers. The second part focuses on the challenges of genetic improvement and seeddissemination for seven crop species of broad importance to African agriculture.

8 Chapter 1


2 The Challenge

2.1 Overview

The exploration that gave rise to this book tried, to the extent possible, to take a cleanslate approach to understanding the role of crop genetic improvement in African agri-culture, and recognize those factors which seemed most influential in how small scalefarmers take advantage (or fail to take advantage) of improved crop varieties. An impor-tant sub-theme of this study was to understand why the Green Revolution of Asia andLatin America did not have a greater level of impact in Africa.

Inevitably, the complexity inherent in the range of factors (farmer income, profita-bility, infrastructure, education, environmental factors, institutional factors, etc.) whichaffect crop improvement in Africa obliged the authors to group some of those factorswhich were perceived as less important tinto those which were believed to be of major,continent-wide importance. The result is a short list of interacting factors whichincludes:

the range and intensity of biophysical constraints to crop growth; large agro-ecological variation; the under-developed state of seed sectors in most countries; the absence of policies which encourage crop improvement; and, very low and declining soil fertility in much of Africa.

While depicted here as constraints, the chapter largely tries to communicate a mes-sage of optimism that previous barriers to raising agricultural productivity in Africa canbe overcome through new knowledge, new science and better methods of working withfarmers.

9

2.2 A Myriad of Production Constraints

The African continent south of the Sahara is dominated by agriculture. Approximately70% of Africans live in rural areas and an estimated 50 million families derive theirlivelihood from farming. The vast majority of these farms cover an area of less than 5 haand are hand-tilled. Crops are grown using a minimal amount of purchased inputs(i.e. seed, fertilizer, etc.) (Wiggins, 2000).

Under these conditions, African crops are threatened by a daunting array ofdebilitating production constraints which farmers can do little to change. In this book,these constraints are loosely categorized as either routine or intractable. Routineconstraints are those which may be more or less effectively controlled through plantbreeding aimed at raising genetic resistance or tolerance levels through conventionalcrossing and selection methods. Intractable constraints are those which are difficult orimpossible to control through conventional crop improvement (see Plate 2).

Categorization of a constraint as either intractable or routine is of course dependentupon the ability of the farmer to alter the growing environment. The very limited invest-ment capacity of small-scale farmers in Africa means that many potentially routineproduction problems are, in fact, intractable. This increases the significance of geneticcrop improvement as a strategy in their potential control. Likewise, the dominanceof production constraints shifts the breeding strategy from one aimed at maximumyield potential under high input use to one aimed at limiting losses from identifiedconstraints under low input use.

The potential to manage production constraints through crop genetic improvementhas increased steadily throughout the history of plant breeding, but has been greatlyexpanded through the emergence of biotechnology. Although the diverse applications ofbiotechnology may eventually make it a useful approach to the control of routine con-straints to food production in Africa as well, for the purposes of this book, biotechnol-ogy is considered primarily applicable in the case of intractable constraints. Anincomplete, but illustrative short list of intractable constraints to production for sevencrops in significant portions of Africa is given below (Table 2.1).

While constraints to crop production exist throughout the world, they are moreintense in the tropics. Wellman (1968) studied the incidence of diseases on a number ofimportant food crops and noted far more in the tropics than in temperate areas. Doverand Talbot (1987) reported that preharvest losses due to pests and diseases are approx-imately 3550% in some tropical areas (Table 2.2).

Biophysical constraints in Africa pose a greater threat to increasing agriculturalproductivity than in other developing regions of the world. African farmers use vastlyfewer off-farm inputs and largely continue to apply traditional methods of cultivation(Wiggins, 2000). In contrast, Latin American and Asian farmers have broadly modern-ized their cultivation methods over the past three decades (Table 2.3).

Table 2.3 indicates the vastly contrasting pace of development in different regionsin the developing world over the previous three decades. The agricultural sectors of Asiaand Latin America, which began the period with higher levels of development in allcategories (irrigation, fertilizer use and mechanization), have developed more rapidlythan Africas.

Latin America maintained a fourfold advantage in the percentage of irrigatedagricultural land over Africa. Asia, which began the period with a massive, 24-fold

10 Chapter 2

advantage in the percentage of irrigated land, continued to add irrigated land at a rapidpace, while Africa, continued to grow from a miniscule base.

Significant variations in rates of growth are also noted in fertilizer use and mech-anization. Asia and Latin America finished the three-decade period with more thanfivefold and threefold increases in fertilizer application rates, respectively, while African

The Challenge 11

Focus crop Intractable traits

MaizeSorghumMilletRiceCowpeaCassavaBanana

Striga, stem borers, phosphorus uptakeStriga, anthracnose, phosphorus uptakeStriga, head miner, downy mildewGall midge, rice yellow mottle virusMaruca pod borers, bruchids, thripsRoot rots, green miteBanana weevil, nematodes, black sigatoka

Table 2.1. Examples of intractable constraints to production among small-scalefarmers for seven important African food crops.

Number of diseases

Crop species Temperate areas Tropics

Sweet potatoRiceBeansPotatoMaize

1554529185

187500600253280

175125

Source: Dover and Talbot (1987) after Wellman (1968).

Table 2.2. Crop disease incidence in tropical compared with temperate zones.

Irrigated area(% of total

agricultural land)

Fertilizerapplication(kg ha1)a

No. of tractorsin use( 103)

1970 1997 1970 1997 1970 1997

Sub-Saharan AfricaAsian developingcountriesLatin America andCaribbean

0.4

9.6

1.5

0.6

13.4

2.4

1.2

9.0

4.4

2.9

52.5

14.6

84

488

637

159

4610

1589

Source: FAO (2000).aFigures vary. These were calculated by dividing FAO total fertilizer consumption bytotal cultivated area.

Table 2.3. Rates of usage of irrigation, fertilizers and mechanical land preparationin Africa, Asia and Latin America.

farmers managed only an increase of between two- and threefold. African farmers nowapply fertilizer at lower rates than Asian and Latin American farmers did three decadesago. Lower input use in Africa is probably substituted in part by added labour input,without which, yield levels would be lower. This translates to lower labour productivity,and an accompanying drag on management capacity within the household, withresultant negative impacts on factors such as sanitation, education and infant health.Even more striking differences were noted for mechanical land preparation. In 1998,Africa had one-third and one-quarter, respectively, the number of tractors in use as Asiaand Latin America in 1970.

All these factors irrigation, fertilizer and mechanization exert a homogenizingforce on crop growth conditions when they are present. Irrigation (and drainage) makeswater more uniformly available to the plant throughout the season, allowing for theplant leaf canopy to remain fully extended over the full growing cycle. Irrigation alsosignificantly reduces risks associated with other forms of investment, such as fertilizers,which fail to provide a cost-effective response in the absence of water. Fertilizer applica-tion, in addition to supplying basic nutrition for the development of vegetative andreproductive structures, reduces variability of nutrient supply within the field, generallyincreasing the value of genetically uniform crop varieties. Tillage performed by tractorsreaches deeper into the soil profile and, over time, reduces localized variation in thefields topography.

The effect of input use is both a whole-farm environment that is more favourableto crop growth than the surrounding, natural environment, and reduced within-farmvariation. It is in part the reduction of this within-farm variation that makes possible thecultivation of highly uniform varieties of a single crop species possible throughout largeareas of North America, Asia and Latin America. As we will attempt to demonstrate, thesame is not true in Africa.

The preponderance of production constraints among African staple crops callsfor increased funding for research on crop genetic improvement to overcome thoseconstraints. While biotechnology is not an automatic solution to these constraints,it should be viewed as a useful tool for improved food security in Africa. Tissue culturecan assist in the rapid multiplication of pathogen-free and true-breeding lines. Geno-typic analysis through marker-assisted breeding can be used to identify favourableindividual plants with valuable, difficult-to-measure traits. Gene transfer throughgenetic engineering can overcome limited genetic variation within a given species.However, as emphasized throughout this book, biotechnology research should be linkedfrom the beginning to viable field-based breeding programmes, and, ultimately, to seeddissemination strategies to prevent their results from remaining on-the-shelf.

2.3 Africas Diverse Cropping Landscape

Africas cultivated area is of immense size and has great environmental variation. Africanfarmers have developed complex cropping systems to fit environments ranging from theslopes of Mt Kenya to the fringes of the Sahara, each with its unique mix of biotic andabiotic constraints. For this reason, cropping patterns and dietary staples vary widelyfrom one end of the continent to another. Moreover, observations of small- comparedwith medium- and large-scale farmers in Africa show that small-scale farmers tend to

12 Chapter 2

cultivate a wider range of crop species, most likely as a strategy for maintaining house-hold food security during a maximum portion of the year independent of householdpurchasing power. The need for diversification may drive farmers to cultivate a verywide range of crop and animal species (see Conway (1997) for an example from westernKenya). In isolated regions of the continent where species diversity is limited, intra-species (or, varietal) variation may be substituted. Farmers in Sudans Bar el Gazahlregion cultivate up to four varieties of sorghum whose morphological and growth habitdifferences rival those found among different crop species.

Farmer crop deployment strategies extend to the species and subspecies levelaccording to complex environmental and social norms. Rice farmers in northern Maligrow large plots of relatively high-yielding Asiatic (Oryza sativa) rice on upper-levelterraces of their farms on the Niger flood plain for normal consumption during the year.In lower-lying parts of their farms, they grow preferred, African (O. glaberrima) varietiesfor use mainly during special occasions such as religious holidays, weddings, andbaptisms. Farmers in the Bugusera region of Rwanda and Burundi grow differentvarieties of sorghum and bananas in the same fields for use in either beer-making or asweaning foods for infants. Farmers in a wide range of agro-ecologies of eastern andsouthern Africa grow small plots of sesame as a source of cooking oil, which otherwiserepresents a major household cash expenditure.

The interaction of opportunities and constraints that farmers manage creates theresultant farming systems that embody the use of all available resources human,ecological, genetic, and other for achieving food security. Researchers have at variousjunctures attempted to understand this full picture of the farming system throughextensive interaction with farmers prior to intervening through research initiatives(Hildebrand, 1981; Merrill-Sands, 1986).

The complexity of farmers decision-making environments can be startling. Table2.4 shows the agricultural calendar prepared by farmers and extension agents in TeteProvince, Mozambique (Buhr, 1990). A tremendous amount of information can beinferred from the chart, which depicts, for example, what some agricultural economistshave long asserted about African farming systems, namely, that labour is often aconstraint in modifying existing cultivation practices (Barker and Cordova, 1978;Hildebrand and Poey, 1985). While labour shortages are obviously apt to exist duringthe September to November planting period, additional shortages can occur duringmuch of the rest of the year, as well, including during the time of weeding and harvest,when certain niche crops and second seasons (or relay crops) of main crops mustbe planted. These labour shortages often lead to late planting in large parts of Africa.This, combined with the existence of a recurrent hunger period prior to the mainharvest season, gives rise to the intense interest farmers in Mozambique andelsewhere have shown in earlier-maturing maize. Early-maturing varieties can also leadto the introduction of a second relay crop, which can be grown on residual moisture,thus permitting a broad intensification of farming systems (Haugerud and Collinson,1990).

While such complexity can appear overwhelming to researchers attempting to makecontributions through the transfer of improved technologies, experience suggests that itis just this level and type of information that is needed in targeting different agro-ecologies from a limited number of research sites, as in the case of plant breeding. Theseprincipals are explored in greater detail in Chapter 4.

The Challenge 13

14C

hapter2

pea

TheC

hallenge15

pea

Okra

Because small-scale farmers in Africa cultivate largely unimproved (i.e. unterraced,non-irrigated, and undrained) fields, they have deployed a rich mix of crops, eachbearing an adaptive advantage within some niche on the farm. The implication is thatfarmers of a given agro-ecology seeking to improve their overall productivity may be inneed of improved, adapted cultivars of several crop species. Likewise, significantlyimproving any one crop may be of great benefit to one group of African farmers but oflittle importance to another (see Plate 3).

While household economies and farm-level variability create crop deploymentpatterns at one level, large-scale environmental variation on the African continentcreates crop deployment trends on a far wider scale. Understanding these trends can beof use in devising strategies for agricultural research at national and regional levels.Figure 2.1 shows the geographical distribution of production of staple crops within themajor climatic zones of Africa.

As would be expected, the distribution of crop species across differing geographicalregions varies considerably. Within a region, however, priorities can be relatively easilyidentified. In order to significantly affect household food security status of ruralpopulations, efforts focusing on a given region obviously must focus on a range of cropsused extensively on farms within that region.

16 Chapter 2

Fig. 2.1. Per capita production (in kg) of primary crops in sub-Saharan Africa. Ma,maize; R, rice; S, sorghum; Mi, millet; Ca, cassava; B, banana; Co, cowpea. Source:FAO (2000).

A major impact on food security in Africas semi-arid sahelian zone will requireinclusion of drought-tolerant crops such as sorghum, millet and cowpea.

For humid, lowland regions such as coastal West Africa and the Congo Basin,strategies should include a focus on cassava, with important, secondary efforts onmaize and rice.

Sorghum and millets, formerly major crops in eastern and southern Africa, arelosing acreage and would appear to have decreasing importance in this region.Today, strategies for most rural populations of eastern and southern Africa need tofocus primarily on maize, with important, secondary efforts on cassava, commonbean and banana.

Grain legumes remain important in the diets of people in all areas except the CongoBasin, although their inherently lower yield prevents them from competing withproductivity levels of starchy crops. This translates to a focus on cowpea in lowlandareas and common beans in mid-altitude and highland areas.

Dealing with environmental variation requires a strategy which encompasses the geneticchallenge (namely, introgressing target traits useful to a range of crop species into a rangeof crop genetic backgrounds) (Buddenhagen and de Ponti, 1983) and the institutionalchallenge (working through structures which link differing national programmes facingsimilar crop improvement tasks). However, through better understanding of theprioritization given by farmers to different species and traits within those species, itshould be possible to develop coherent, national strategies for improving the geneticbasis for crop production for a range of species.

Species selection by farmers across various subregions is reflected in consumptionpatterns for Africa as a whole. Figures 2.2 and 2.3 show figures for per capita consump-tion of the principal food crops in Asia and Africa during 1997. The data reveal the useof a wider range of food crops in Africa. Consumption of only two crops rice andwheat accounts for 70% of non-animal food consumption in Asia. Meanwhile, theconsumption of Africas four most important crop-based food products wheat, maize,banana/plantain and cassava accounts for only 67% of its total, with much of thewheat being imported.

The broad implications for crop improvement strategies in the two regions areobvious. Whereas Asias struggle largely hinged on the ability of researchers and farmers

The Challenge 17

AsiaAfrica

9080706050403020100

Crop

con

sum

ptio

n(kg

perso

n1 ye

ar1 )

Maize

Whea

tRic

e

Sorgh

um Millet

Fig. 2.2. Cereal crop consumption trends in Asia and Africa, 1997.

to devise more productive rice and wheat-based farming systems, in Africa, broad-basedfood security will require sustainable productivity increases within its respective agro-ecological systems based on maize, sorghum, cassava, millet, rice, pulses and bananas,among other crops. Consequently, this book focuses on seven priority food crops ofsub-Saharan Africa. In Asia, the initial success of modern varieties of irrigated rice andwheat was followed by success in developing varieties for many rain-fed areas devotedto these and other crops. Most of Africa, however, is characterized by farming con-ditions that have reduced or delayed the impact of genetic improvement found inother parts of the world. While all the major regions of the world include areaswith these sets of conditions, in Africa, they dominate. As indicated by Byerlee (1996),these include marginal crop production areas, areas with very poor infrastructure, andareas where quality traits outweigh the yield advantages of improved varieties, amongothers.

An observational accounting of land use patterns among small-scale farmers inAfricas little-modified agricultural landscape indicates that crops are employed largelyaccording to their ecological niche. Issues such as temperature, natural drainage, rainfallpatterns, soil fertility, and pest and disease occurrence, to a large extent, govern whichcrops can be used where. In the tropics, the temperature regime is mainly influenced byelevation. Most attempts at classifying cropping systems have focused on this factor. Inthe following, an attempt is made to offer descriptions of the trends in agricultural landuse in differing environments in Africa.

Lowlands. Valley bottoms of lowland humid environments are widely sown with rice. Inoff-seasons, raised beds often produce sweet potato. In sloping areas, cassava and uplandrice are grown on highly leached soils. As rainfall decreases, soil phosphorus levelsincrease and maize can be grown. Semi-arid lowland zones are dominated by sorghumand millet cultivation. Cowpea is the most important pulse crop grown in all well-drained lowland environments. Few pulses grow well in poorly drained lowlandenvironments, and diets often lack this element.Mid-altitude zones. Mid-altitude zones of Africa are dominated by maize. In higherrainfall areas, however, maize productivity is reduced by foliar and storage pests anddiseases and reduced sunlight, and cassava and/or bananas are commonly grown.

18 Chapter 2

AsiaAfrica

80

70

60

50

403020100

Crop

con

sum

ptio

n(kg

perso

n1 ye

ar1 )

Cassa

vaPo

tato

Swee

t pota

toYa

ms

Bana

na/pla

ntain

Pulse

s

Fig. 2.3. Consumption trends of selected non-cereal crops in Asia and Africa, 1997.

Cassava is also substituted for maize in very high population zones and areas with verypoor soils. In lower rainfall areas, moisture stress reduces yields and sorghum becomesthe dominant crop. Poorly drained mid-altitude environments are planted to rice. Beansand pigeon pea are the most popular pulses in mid-altitude zones.Highlands. Areas above 2000 m except Ethiopia are planted to English potato andhighland bananas, with interspersed plantings of maize and wheat. In Ethiopia,highland areas are planted to teff. The dominant pulse of the African highlands is beans.In the Great Lakes regions, beans also supply a large percentage of carbohydrates.

The result is a rich and resourceful utilization of crop genetic resources which con-tribute economic advantages, nutrition, and cultural significance to rural householdsacross the continent. Moreover, this diversity, while presenting its own challenges tocrop improvement, holds the promise that improvements made on crop species can beused in ecologically sustainable ways. As explored in greater depth later in this book,both the variety of crops and the importance of their adaptation, in turn, highlightthe need for decentralized breeding operations and the continual involvement of farm-ers in identifying traits and selecting improved crop varieties for multiplication andcommercialization.

Regional crop improvement programmes have made some attempts at under-standing the complexity of agro-ecologies in Africa in order to target better their breed-ing efforts and varietal testing programmes (see Plate 4). The International Center forMaize and Wheat Improvement (CIMMYT), for example, has recognized nine maizeproduction mega-environments in sub-Saharan Africa based on three different altituderanges in three different ecologies: lowland tropics, subtropical, and highland tropics(CIMMYT, 1990). The International Center for Tropical Agriculture (CIAT) hasrecognized 14 bean production environments in sub-Saharan Africa based on similarcriteria (Wortmann, 1998). Such groupings of environmental variation into aggregatedgeographical units is critical for the targeting of crop genetic resources aimed atachieving continent-wide coverage. However, the level of resolution achieved by suchefforts to date remains imprecise in comparison with its importance in hitting the targetconsistently, throughout Africa. Thus, crop-specific agro-ecological analysis remains acritical area of untapped potential for broadly improving the impact of crop geneticimprovement.

In most cases, this will be a task taken on by the national agricultural researchsystems (NARSs), at times reinforced by ecological and geographic informationsystems (GIS) studies performed by international agricultural research centres (IARCs)outreach programmes. Progress has been made in several countries. As an example, theKenya Agricultural Research Institute (KARI) recognizes five maize breeding agro-ecologies in Kenya which are used to focus breeding efforts, depicted in Fig. 2.4.Researchers in western Kenya have recognized additional subclassifications within themoist mid-altitude zone, which can be used to add further precision to crop improve-ment efforts (Amadou Niang, personal communication). Thus, even cursory inventoryof maize agro-ecologies in Kenya may result in six or more broad families of maizevarieties. Kenya represents one of the most intensively studied countries in Africa, and,perhaps not coincidentally, one where crop genetic improvement has made significantimpact (Gerhart, 1975). Wider and more intensive analysis of agroecologies needs to beconducted by crop improvement teams in all African countries.

The Challenge 19

Trying to make accurate determinations of varietal needs for numerous groups ofAfrican farmers living in widely varied agricultural agro-ecologies is a real challenge.Without it, however, the chances of success are slim. Experienced crop improvementspecialists in Africa can cite the many new varieties which have been developed for

20 Chapter 2

Fig. 2.4. Constraints to maize production in major agro-ecologies (Highlands, Mid-AltitudeMoist, Mid-Altitude Intermediate Moist, Mid-Altitude Dry, Mid-Altitude Intermediate Dry, andTropical Lowlands) of Kenya identified by KARI maize breeding teams. DTM, days to maturity.

African farmers, but which have never been adopted. Former US Secretary of Agri-culture, Clayton Yeutter, writing recently on the biotechnology revolution (ISAAA,2000), stated:

Newer genetic modifications, impressive as they may be in the laboratory and in the pagesof professional journals, are of little real world relevance unless those desirable traits aretransmitted through seeds with good yield characteristics. Otherwise, farmers in the U.S.,Africa, or anywhere else, simply will not plant those crops.

Unlike Asia, new crop varieties for Africa cannot be developed based on the assumptionthat fertilizer will be subsidized and made available through government programmes.To be adopted in Africa, new varieties need to be well adapted to local conditions andprovide yield advantages with few external inputs. Recent approaches to breedingfocused on selection under low-input African conditions (Bnziger et al., 1997;Bahia and Lopes, 1998) have proved effective in identifying varieties with superiorperformance under drought and low soil nutrient status. Such adaptation to environ-mental stress needs to be combined with good levels of resistance to foliar diseases, insectpests and, in some cases, the ability to grow vigorously during early stages of develop-ment to shade out weeds. While landraces will in most cases have reasonable levels ofresistance to all these constraints, for reasons related to co-evolution and the relativelyslow rate of genetic change via mass selection methods performed by farmers, it isunlikely these levels will match those possible through scientific breeding programmes.

To develop varieties that poor farmers find useful, it is necessary to understandenvironmental variation in Africa and listen to farmers advice on issues of growingenvironments and household utilization, a topic explored more extensively in Chapter4. In Africa, perhaps more than in any other part of the world, the science of geneticimprovement must be paired with the art of understanding people and the environ-ment. This will require additional investment in areas which serve to consolidate thepresently diffuse, dispersed base of knowledge on agro-ecologies and crop user systemsin Africa. Recent initiatives such as the atlases on bean, cassava and maize are a good startin this direction (Carter et al., 1992; CIAT, 1998).

Household preferences, as well, cannot be overlooked in breeding programmes andconsideration should be given to the overall crop usage environment in which theadoption must take place. Some crop/user system combinations in developing countriesconstitute situations where yield advantages of improved varieties can easily be out-weighed by the importance of quality traits (Herdt and Capule, 1983). Very poorfarmers often cannot afford to pay for industrial milling services, and must carry out allprocessing tasks in the home. Thus, farmer preference for flint-textured maize varietiesamong resource-poor farmers in Malawi was key to identifying flint hybrids whichachieved high levels of adoption in the early 1990s (Nhlane, 1990; Smale et al., 1993).Likewise, food scientists who have analysed sorghum quality characteristics have becomeincreasingly capable of predicting the acceptability of improved varieties based on thequality of food products they produce. Studies conducted using sorghum flours fromWest, southern and East Africa revealed significant differences in flour texture and totalwater content of porridges consumed. Households preferred varieties with high amylosestarch content and low flour lipids and proteins (Fliedel and Aboubacar, 1998). Fewimproved varieties have scored high in such tests. Nevertheless, breeders have often

The Challenge 21

failed to take full advantage of the ability of food scientists or consumers to inform themof the probable success of their offerings at the household level.

The need to link as much field-based information as possible to crop improvementprogrammes argues for a high degree of integration of disciplines and connectivitybetween breeders working at international, regional and national levels. Decision-making matrices that may seem very complicated to breeders may be a relatively simplematter for farmers who use crops in various forms everyday. Since many aspects ofadaptation and farmer preference do not relate to expertise commonly embodied withincrop research institutes, adequate linkages need to be established with agencies orindividuals who do embody this expertise, including farmers and NGOs.

2.4 A Seed Sector Dominated by Market Failure

While the trend toward privatization and globalization of the germplasm sector hasundoubtedly resulted in the distribution of better seed-based technologies to farmers indeveloped regions of the world, these policy changes will not function to the same extentin Africa in the short or medium terms. Private seed companies are constrained tooperating in environments where they can make acceptable profits. In Africa, multi-national seed companies may be motivated to popularize one or even several high-yielding maize hybrids among better-off farmers in favourable areas, but it is less likelythat they will find it profitable to devote significant resources to developing varietieswith the very specific adaptation advantages required by small-scale, low-input farmers.Even if such varieties enabled resource-poor farmers to double their yields, this wouldoften mean an increased harvest of only 1 t ha1 or less. The share of increased profits aseed company might capture from such a modest increase is small in comparison withprofits available in developed regions of the world (Tripp, 2000). Additionally, thedegree of complexity involved in designing a full range of varieties required by differentcategories of farmers cultivating farms in very different agro-ecologies further limitspotential profitability.

Low effective demand and relatively small profits available from seed in much ofAfrica, in comparison with the rest of the world, have delayed the commercialization ofthe seed market. Low rates of economic growth forecast for much of Africa are not likelyto attract large-scale investment from outside the continent of the type needed to achievebroad coverage of farmers needs for seed. Rather, indigenous seed companies that oper-ate closer to local markets and on lower margins should be considered as a solution withwider potential. To date, however, little international or national assistance has beendirected at this type of company. The strategy put forward in this book places highimportance on the development of Africas private seed companies.

Regardless of the strategy employed, given the economic realities in Africa and thedifficulties seed companies face in attracting clientele, growth of the seed sector is likelyto be slow and sporadic. The implication is that public sector-based strategies for seeddissemination will be critical to realizing the benefits of crop genetic improvement inAfrica for some time to come. In fact, the absence of a sufficient effort by either privateor public sector breeding interests has left an enormous gap in the seed supply offered toAfrican farmers. While things can be done to encourage such investment, alternativestrategies and continued experimentation are needed (see Chapter 6).

22 Chapter 2

Finally, in the absence of investment by private seed companies, the rapid-fire,signalresponse, product-refining process that is the great advantage of distributionsystems conducted via private enterprise will not operate for seed in Africa. Feedback onperformance and preference issues must be gathered through other means. This factdrastically increases the need for continual participation by farmers in variety refinementand seed dissemination. It also points to a critical role for research managers who overseecrop improvement initiatives and are ultimately responsible for ensuring that usefulvarieties emerge from such efforts. This challenge is discussed in greater detail below.

2.5 Policies and Institutions are Critical to the Success of CropImprovement

Policies that favour food security like those which favour education and health provide a foundation for development and the passing on of these basic human needsto successive generations (Sen, 1981). Of these three commonly cited priorities fordevelopment, however, the most basic and immediate is security against hunger. InAfrica, where none of the three needs has as yet been broadly secured for society, equityarguments can be advanced that food security ranks as the most essential priority. Yetpublic and donor funding for agriculture has lagged far behind other priority sectors inAfrica. A recent review of public spending in Uganda showed that agriculture accountedfor just 4% of total expenditures from national and donor agencies sources, comparedwith 11 and 20%, respectively, for health and education (Uganda Ministry of Finance,2000).

Few African countries have prioritized food security through development of theagricultural sector. While speeches made by African leaders are invariably peppered withreferences to freedom from hunger and development of the economy through tech-nological advance, public sector spending on agricultural research and extension inAfrica declined from 1981 to 1991 (Pardey et al., 1997). African governments havereceived little real encouragement to develop their agricultural sectors from Westerngovernments, several of which have de-emphasized the agricultural portfolios of their aidpackages to Africa during the 1990s. The US government made payments to farmers of$7.3 billion in 1995 and 1996 (USDA, 1998). Meanwhile, the prevailing belief is thatthe agricultural sector in Africa should develop itself.

Thus, a fifth challenge is situated within the institutional framework of crop geneticimprovement: how individuals, groups, and institutions are organized to achieve resultsin relation to goals which require collaborative arrangements, resulting in a physicalproduct which is usable by farmers. Overall institutional or departmental performanceinfluences significantly the output of breeding teams and is generally unrelated to theacademic preparation of the individuals involved. The area of public policies andinstitution performance attains greater importance in relation to the many regulatoryissues and intellectual property rights attached to techniques and products ofbiotechnology.

At a national level, there is a need for crop-based strategies for genetic improvementthat make use of the full range of scientific capacity which can be applied. Nationalbreeding programmes are the front lines of public sector breeding in Africa. For many ofthe self-pollinated crops, and for open-pollinated varieties (OPVs) of cross-pollinated

The Challenge 23

crops, national programme varieties are likely to continue to be the dominant means bywhich genetic improvements move out to small-scale farmers in Africa. The efficiency ofthis process is highly sensitive to the science policy of each institution and the regulatoryframework governing plant varieties.

National and international policy on intellectual property and plant varietyprotection is being debated in various corners of the globe, and outcomes are difficult topredict (Barton, 1998; Erbisch and Maredia, 1998; Koo and Wright, 1999). Indeed,confusion exists in many cases regarding what forms of biological property can beprotected where and for what purpose. By some interpretations, the trend would seem tothreaten access by developing countries to genetically engineered crops and, quite possi-bly, to other biotechnology applications as well, as donor agencies reduce support for allbut conventional science applications. The widespread patenting of breeding materialsby both private companies and public universities in the USA and Europe has inevitablyreduced the flow of germplasm from those regions to Africa. However, even at this earlystage, progress has been made towards sharing critical intellectual property that wouldseem to counter that argument (Mugo, 2000). What seems clear is that developingcountries require advocates who work in the interest of broadening access by them toemerging technologies, and not simply the products of those technologies.

2.6 The Soil Fertility Problem

Better varieties, of course, are only part of the challenge. Of equal or greater importanceto realizing the full productive potential of crop plants is the supply of plant nutritionthrough healthy, fertile soil. Traditionally, major increases in productivity have mainlycome about as a result of a combination of improved production systems (irrigation,drainage, improved cultural practices, introduction of fertilizers) and the introduction ofmore efficient varieties (Matlon and Adesina, 1991). The introduction of improvedwheat and rice varieties in Asia, for example, coincided with wider availability ofinorganic fertilizers and irrigation (Herdt and Capule, 1983). Cheap and widelyavailable inorganic fertilizer in Nigeria facilitated rapid expansion and intensification ofmaize production following the introduction of improved, adapted varieties in the1970s (IITA, 1995). More recently, improved maize varieties and increased fertilizerapplications (encouraged by credit facilities) in Ethiopia during the period 1994 to 1996produced a dramatic, 31% increase in average yield (Quinones et al., 1997) (see Plate 5).

Fertilizer consumption in Africa has stagnated, even while land brought undercultivation has increased dramatically. From 1970 to 1982, there was a slight but steadyincrease in fertilizer consumption in Africa, but there has been no increase in totalconsumption since then (Fig. 2.5). Meanwhile, the continent has added 270 millionpersons. The reduced fertilizer use per capita has been made up through bringingnew land into cultivation and by mining soils of their nutrients through continualcultivation without replenishment of nutrients. As a result, net nutrient outflows peryear in Africa are estimated at 63 kg ha1 year1 on average (Debrah, 2000). Lower inputuse in Africa is probably substituted in part by added labour input, without which, yieldlevels would most likely be even lower.

Herein lies an enormous problem. Farmers in sub-Saharan Africa use by far the leastamount of fertilizer in the world. In 1993, average application of total nutrients in Africa

24 Chapter 2

was 10 kg ha1, compared with 83 kg ha1 in other developing regions (Heisey andMwangi, 1996). Moreover, subsidies which formerly encouraged the use of fertilizerswere largely removed starting in the early 1980s, and have not been re-applied. Thehalt in increase of fertilizer applications in Africa shown in Fig. 2.5 coincides nearlyexactly with the implementation of those policies. Studies conducted by Holden andShanmugarathan (1994) and Bumb and Baanante (1996) both showed that higherfertilizer prices following the removal of subsidies led to reduced application ofinorganic fertilizers in several African countries.

Today, domestic fertilizer prices in Africa are far above world prices. While worldurea prices in 2000 ranged between $80 and 100 t1, urea prices in several Africancountries range from $400 to $842 t1 (Debrah, 2000). At prevailing fertilizer costs andfarm-gate prices for commodities, the economics of fertilizer use are not favourable.Extensive analysis of fertilizer responses, fertilizer prices and producer prices for maize inMalawi from 1994 to 1997 resulted in researchers recommending zero application offerilizer on maize produced for market in 34 out of 41 agro-ecologies (Benson, 1997). Insummary, therefore, at current fertilizer prices in Africa, there is little perspective forfertilizer application among African farmers to increase (Sanchez et al., 1997).

Privatization of agricultural input markets has removed much of the flexibility gov-ernments formerly availed themselves of in encouraging the use of agricultural inputs,including fertilizers (Cromwell, 1996). To an increasing degree, therefore, poor farmersare left with fewer options.

The response among soil fertility researchers to reduced applications of inorganicfertilizers has been to focus on the cycling of nutrients in low-input systems and the useof lower-cost methods of adding nutrients, such as legume rotations, green manures,and improved fallows (Sanchez et al., 1997). While significant amounts of nitrogen canbe added to soils through the cultivation of green manure crops, the limitations ofthis method for maintaining soil fertility must not be overlooked. Crop recovery ofnitrogen contributed by the leaves of leguminous plants is generally lower (1030%recovered) than that contributed by inorganic fertilizers (2050%) (Palm, 1995).More importantly, legume biomass contributes little of the phosphorus required tocomplement the nitrogen and potassium contributed by such additions. For example,

The Challenge 25

Fig. 2.5. Total fertilizer consumption in Africa (excluding Egypt and Libya). Source:IFA (2000).

cover crops of velvet bean (Mucuna pruriens) and Crotolaria ochroleuca contribute onaverage 3542 kg nitrogen and 79 kg potassium per tonne of biomass, but only1.62 kg of phosphorus (Palm et al., 1997). A substantial crop of either of these greenmanures of, say, 5 t ha1 would thus yield a maximum of 10 kg P ha1 far below theminimum required for a 2 t ha1 crop of maize.

Humankinds dependence on the HaberBosch process of synthesizing nitrogen foruse in producing its food requirements has been extensively analysed by Smil (1991),who concluded that no alternatives to the use of inorganic nitrogen currently existfor densely populated developing countries. While many African countries have lowabsolute population:land ratios, Binswanger and Pingali (1989) revealed that in reality,due to highly uneven resource endowment in Africa, many countries have high effectivepopulation densities. Heisey and Mwangi (1996) have also described many Africancountries as land-scarce. These considerations, coupled with the recognized labourconstraints many small-scale farmers operate under in Africa, argue for continued effortsaimed at making fertilizers broadly more accessible to small-scale farmers. In an analysisof the factors leading to low fertilizer usage among small-scale African farmers, Debrah(2000) identified as critical factors: high fertilizer cost, low farmer profitability, unequalaccess, low credit availability, and socio-economic factors related to farmers attitudestoward fertilizer use.

The Sustainable Community-oriented Development Programme (SCODP) is anNGO aimed at improving access to fertilizer and seed among farmers in Siaya District ofwestern Kenya. The organization promotes the use of these inputs and improved cropmanagement practices through 11 small farm input shops located in rural marketcentres. By breaking large (50 kg) bags of fertilizer into units ranging from 100 g to 2 kg,the organization became the largest supplier of fertilizer in the district in only 4 years.Moreover, the organization was able to operate at a profit, and projected growth in salesof over 400% by 2002 (SCODP, 2000). Such initiatives SCODP is supported by sev-eral donor agencies may reveal how fertilizers can be supplied at more affordable pricesthan at present. Nevertheless, the scale of such activity remains very small. Only 10% offarmers in Siaya currently use fertilizer, and fertilizer application rates in the SCODPproject area remain low. Even if SCODP were to achieve projected growth in sales, aver-age fertilizer application rates in the District would remain at 6 kg ha1.

Finally, the case of broad provision of small quantities of fertilizer to Africanfarmers deserves attention. Between 1999 and 2000 in Malawi, the Starter-Packprogramme distributed 2.86 million packages to 2.4 million farm households (indicat-ing full coverage but some errors in distribution as well). Packs contained a range of cropspecies, depending on the agro-ecology in which they were targeted, but generallyconsisted of 2 kg of cereal crop seed, 2 kg of legume seed, and 10 kg of fertilizer. Averagehousehold harvest increased from 1087 kg ha1 to 1904 kg ha1, resulting in an averageincrease of 96 kg per household (Mann, 1999). Distribution of the packages boostednational maize production by 25% and extended household food sufficiency from 6.1 to8.7 months (Levy et al., 2000). However, the sustainability of this programme is alreadybeing challenged.

Projected low rates of economic growth coupled with population increases insub-Saharan Africa during the coming decade (Conway, 1997) carry important implica-tions for technology development and application among small-scale farmers. If, as pro-jected, fertilizer applications in Africa are likely to remain low for the foreseeable future,

26 Chapter 2

the primary contribution from plant breeding will come from efforts aimed at makingcrop plants more productive and dependable within low-input, limited-infrastructureproduction environments. Improved nutrient management based on accessible, practi-cable methods of returning nutrients to the soil can make land more productive, andmore resilient crop plants can help produce better harvests; however, it is likely thatAfricas crop production environment will generally remain low in fertility.

Low soil fertility and low fertilizer use, combined with a lack of irrigation, very lowpesticide use, and low rates of tractor use in Africa, effectively mean that crop geneticsand improved planting material remain as one of the most effective means by whichAfrican farmers can be assisted. The absence of these productivity-enhancing, agro-ecology-homogenizing factors of production makes the task of crop improvement morecomplex and more difficult in Africa than in other parts of the world. But low input usedoes not reduce the importance of crop genetic improvement in the lives of Africas ruralpoor, rather, it raises it.

2.7 What, Then, is Needed?

Quite obviously, improving all crops for all possible traits in all agro-ecologies in Africais not feasible in the short or medium terms. One obvious difficulty with agro-ecologicalapproaches is in achieving the scientific and technical acuity at all the levels necessarybroadly across the continent. Without making a detailed analysis of the number ofbreeders and other crop scientists currently employed at the national and internationallevels in Africa, it is safe to say that there are currently insufficient numbers to applysystematically agro-ecology-based strategies for crop improvement. However, if oneaccepts that the biophysical challenges in Africa are great, and that new sciencecombined with new methods for understanding and deploying the varieties that farmersdesire does create a new realm of the possible in Africa, then a concerted, worldwideeffort at improving overall agricultural productivity in Africa, including the geneticperformance of Africas food crops, is justified. Moreover, simple calculations ofthe financial requirements for funding the national component of the challenge(shown in Chapter 4) reveal that making this effort is not necessarily prohibitivelyexpensive.

Making use of lessons learned over the past decade of working closer with farmersin Africa, and combining those lessons with breakthroughs made in genetics, thefollowing steps can and should be taken.

Identify traits that genuinely reduce productivity at the small-scale farmer level ina wide range of environments in Africa and for which a (full or partial) geneticsolution may be possible.

Understand and conceptualize the various agro-ecologies within which new cropvarieties need to be deployed, and determine the priority constraints, adaptationadvantages, and user preferences which need to be targeted in developing newvarieties.

Identify individuals and institutions capable of making meaningful contributionsto a genetic solution to each trait.

The Challenge 27

Provide opportunities for interaction between groups of scientists, aimed atidentifying strategies for developing the available genetic resources into a farmer-usable product.

Implement strategies via vertically integrated crop improvement initiatives whichincorporate biotechnology, breeding, and seed systems.

Disseminate new technologies via farmer-focused, agro-ecologically informedinitiatives which consider the full range of agronomic and genetic technologieswhich can benefit farmers.

Due to the complexity and size of the challenge implied, a long-term commitment isrequired. Only through successive research steps, each carried to the next stage, canprogress remain on track and capable of delivering needed products.

Because biotechnology, breeding, and seed systems form relatively discretefunctions within the crop improvement process, they can, to a certain extent, beconsidered as separate activities, each with its own sectoral strategy. They are consideredas such in Chapters 4, 5 and 6 of this book. First, however, it is worthwhile to documentthe level and the kind of food that Africa needs and to understand how improved farmerproductivity might contribute to a solution. Chapter 3 presents a brief look at the natureof food scarcity in Africa.

28 Chapter 2

3 The Roots of Hunger

Rural communities in Africa are under pressure on several fronts. Profits from farmingat the current, low levels of productivity, are too small to allow farmers to reinvest intheir land and maintain sustainable production systems (Eicher, 1990; Blackie, 1994).Meanwhile, continual increases in population have depleted the available resource baseand eroded many social entitlements which hitherto provided for a state of equilibriumin rural areas of Africa (Lele, 1989). Finally, steady increases in agricultural productivityin developed regions of the world (increasingly facilitated by biotechnology), combinedwith persistent payments of massive subsidies to North American and Europeanfarmers, have continued to push world grain prices downward, making it increasinglydifficult for marginal land farmers in developing countries around the world to operateprofitably (FAO, 2000). Rural areas by definition offer a limited set of economicalternatives to agriculture, and Africa has attracted very little direct foreign investmentto create new jobs, even in urban areas. As a result, economic growth in rural areas hasbeen insufficient to offer alternative means of employment for the rural poor (Eicher,1990; Oyejide, 1993), and agriculture remains their only real option for survival andincome.

Africa has the highest percentage agricultural population and the second highestcultivated area in the world (Table 3.1). However, average cereal yields are the worldslowest, and less than half of Asia and Latin America.

Another reason for rising hunger in Africa during the past few decades is high ratesof population growth. In 1970, Asia, Latin America and Africa had similar rates ofpopulation growth (Fig. 3.1). Yet population growth in Asia began decreasing rapidlythereafter and by 1978 had decreased by 25%. Latin American population growth ratesdeclined more slowly, but by 1995 fell by nearly 20%. In Africa, the rate of populationgrowth rose sharply between 1970 and 1995. At present, the population growth rate inAfrica is nearly double that of Asia.

A high population growth rate is especially injurious in countries

DeVries &Toenniessen_ Securing the Harvest.pdf

Documents