cgiar system-wide livestock programme · compared. it should also greatly speed up the development...

CG

IAR

Sys

tem

-wid

e Li

vest

ock

Prog

ram

me

Biennial Report 1999-2000

IFPRI

CIMMYT

ISNAR

WARDA

IITA

ILRI

ICRISAT

CIFOR

IRRI

ICARDA

ICRAF

IPGRI

CIAT

CIP

IWMI

ICLARM00

About the SLP

Poverty, food insecurity and a deteriorating environment threaten the livelihoods andeven the lives of millions of rural people in developing countries. Smallholder farmersin these countries have few resources or opportunities to improve their situation, butone option is widely available: by integrating crop and livestock production, farmerscan improve their farm productivity while protecting their natural resources.

Linking the livestock-related research of its partners worldwide, the System-wideLivestock Programme of the Consultative Group on International Agricultural Research(CGIAR) is a unique vehicle for enhancing the contribution of animal agriculture to theCGIAR’s objectives of increasing food production, eradicating poverty and protectingthe environment.

AddressFor further information about the SLP, contact:

Dr. Jimmy SmithProgramme CoordinatorILRIPO Box 5689Addis Ababa, Ethiopia

Telephone: + 251-1-615-867Fax: + 251-1-611-892/614-645E-mail: [email protected]

Key: SLP participating centres; other centres

See inside back cover for information about the CGIAR and the names of its centres

CGIARSystem-wide Livestock

Programme



ii

ISBN 92-9146-083-4

Correct citation: International Livestock Research Institute (2000). CGIAR System-wide LivestockProgramme: Biennial Report 1999-2000. Nairobi, Kenya.

iii

Preface

Preface

If we can sum up the task of the CGIAR System-wide Livestock Programme in a singlephrase, it is to “think globally and act locally on livestock-related issues”.

Let’s take the action first. In the few years since it was founded, the SLP has become fullyfunctional as a decentralized team of researchers, each based at a different CGIAR centreand each fully engaged in collaborative research at the regional and national levels. Theseresearchers span many disciplines, ecosystems and research themes, making vitalcontributions in such areas as animal nutrition, crop breeding, the maintenance of soilfertility and the testing of new legume species. They span the research spectrum, fromstrategic research on the genomes of major grain crops to adaptive research with farmers onnew forages. Through the new kinds of partnership that have emerged in recent years, theyare able to integrate their work with that of others, learning from them and sharing theirresults with them.

But the SLP team does much more than this. Its “thinking” role is a strategic one in whichthe team seeks to maximize the benefits from the investments made globally in livestock-related research. This it does in several ways. It identifies and fills gaps in existing researchactivities, often supplying the “missing links” that constrain others in their pursuit of impact.It promotes the pooling of resources and expertise among institutions, to create synergiesand spillovers. It seeks to build a livestock perspective into the crop improvement andnatural resource management research conducted by its partners. It looks for opportunitiesto transfer successful technologies and methodologies, both within and across regions.And it supports strategic research on topics and commodities of global importance.

If all this sounds rather abstract, let us illustrate these points from some of the projectssupported by the SLP and described in this report.

The SLP’s gap-filling and technology transfer roles are illustrated by its contributions toresearch in the highlands of Eastern and Central Africa. Here three inputs to increase andsustain impact were needed. The first was to promote the regional diffusion of legume feedtechnology already known to be attractive to farmers participating in on-farm trials. Thisneeds to be done quickly in order to maximize the benefits from research in the near term.The second two inputs were designed to underpin these benefits in the longer term bydiversifying the plant species on which the technology is based and ensuring the maintenanceof soil fertility. All three inputs are contributing to what is rapidly becoming an Africansuccess story.


iv

Pooling resources in a collaborative effort is a well established path to more effective research.The SLP’s task is to ensure that the livestock perspective is fully integrated in multi-partnercollaborative research. In West Africa, the SLP is contributing to the testing of a new, moreholistic approach to on-farm research in which researchers from different institutes designexperiments together, selecting a limited number of technologies that they believe have thebest chances of achieving an impact. The approach has brought the livestock componentof the production system to the fore for all the participants. It promises to deliver benefitsto farmers in a region where success has proved elusive in the past. And, if it works, it willbe extended to other parts of the developing world.

Building a livestock perspective into research on crop improvement and the managementof natural resources is often vital to the success of such research. In crop improvement, theSLP supports a global project on the use of genetic markers to develop improved varietiesof sorghum and pearl millet, two cereals widely grown by resource-poor farmers to providefood for their families and feed for their animals. This project is showing that it may wellbe possible to improve fodder yield and quality without sacrificing grain yield, a prospectnever even envisaged by conventional plant breeders in the past, who concentrated almostexclusively on raising grain yields. To maximize the spillover from this research, the SLPsupports a second project to extend the findings to a range of other grain crops and grasseswith similar genomes. In natural resource management, the SLP is a partner in severalprojects in which livestock provide farmers with the economic incentive to plant resource-conserving legumes. For example, research by Tropileche, a consortium of Latin Americaninstitutions, is demonstrating the environmental benefits of such forage species as Cratyliaargentea and Arachis pintoi, in addition to their feed value.

SLP-supported strategic research to improve the feed value of crop residues covers maize,sorghum, pearl millet and cowpea—crops grown by millions of resource-poor farmersthroughout the developing world. If successful, this research will have a profound andwidespread impact on poverty eradication. The SLP’s portfolio of commodities and researchthemes is, then, of truly global import.

Lastly, let us touch on one SLP project that underpins all aspects of the SLP’s role ofthinking globally. The SLP is building an Internet-based version of itself, consisting of anenhanced electronic communications and information network and a set of on-linelaboratories for conducting collaborative research. By “going virtual” in this way, the SLPis shrinking not only the distance that separates members of a global research team but alsothe high transaction costs of global collaboration. This project should make it easier todesign and implement research so that results from different locations can be analysed and

v

Preface

compared. It should also greatly speed up the development and transfer of results withglobal applicability.

Besides the projects described in this report, the SLP has recently embarked on a numberof exciting new ventures. Among them is a project on the livestock component of rice-wheat systems in the Indo-Gangetic plains of South Asia. Sharpening the focus on livestockin these vast, irrigated lands could provide thousands of farmers with urgently needed waysof diversifying their sources of income. Two other projects cover the mountain ecosystemsof the developing world, a vital resource for livestock production that has been neglectedby research in the past. One will address the needs of the Caucasus, a region in which theCGIAR is stepping up its activities, while the other focuses on the high Andes and theHimalayas in an attempt to transfer solutions between the two ranges, which suffer similarpressures on their fragile environments. A fourth project will seek to improve the integra-tion of crop and livestock production in Southeast Asia, a region where livestock representa possible route out of poverty for millions of rural people affected by economic recession.And a fifth will investigate two socio-economic issues associated with the rising demandfor livestock products: how the rapidly expanding pig and poultry sectors will affect thesupply of grain for human consumption, and whether small-scale livestock producers canstay competitive in the face of rapid growth in large-scale peri-urban enterprises. Added tothe existing portfolio of projects, these and other new initiatives mean that the SLP nowaddresses the full breadth of the global livestock research agenda, providing it with anunrivalled vantage point from which to identify and capitalize on the major opportunitiesfor impact.

The SLP has come a long way in the few years since it was founded. We believe this reporttestifies to its encouraging progress. We are, however, well aware of the considerablechallenges that remain. We count on your support in meeting them.

Hank FitzhughDirector General, ILRI

Addis Ababa, 1 September 2000

vii

Preface

Executive Summary

The System-wide Livestock Programme (SLP) of the Consultative Group on InternationalAgricultural Research (CGIAR) aims to increase the returns to investments in the CGIAR’sresearch by linking livestock-related activities among its different centres and their partners.The SLP conducts or supports research to improve animal feed resources, integrate livestockproduction with the management of natural resources, and identify new livestock-relatedpolicy options.

In 1997, under Phase I of its activities, the SLP began supporting three ecoregional researchprojects, one each in the cool, moist highlands of Eastern and Central Africa, the hillsidesand forest margins of Latin America, and the dry rangelands of the Sahel and West/CentralAsia-North Africa. All three projects have made excellent progress.

In the highlands, researchers have identified new areas suitable for extending the popularlegume shrub Calliandra calothyrsus, at the same time as intensifying their efforts to developalternative fodder species so as to make the production system less vulnerable to pests anddiseases. Research on the use of manure is under way with a view to identifying opportunitiesto improve its contribution to soil fertility. In Latin America, the consortium Tropilechehas achieved considerable success with Cratylia argentea, a protein-rich, drought-tolerantshrub legume that can increase the profitability of smallholder milk production and isproving particularly popular with farmers in Costa Rica. The consortium is also conductingresearch on other legumes, including Arachis pintoi and Stylosanthes guinanensis. In West/Central Asia-North Africa, the potential of currently underused fodder trees and shrubssuch as Atriplex has been demonstrated, while research in the Sahel has revealed that anumber of popular browse species are now in need of in situ conservation as producerscompete for scarce remaining resources to meet the rapidly growing market for livestockproducts. Researchers are assessing the potential for improving the productivity of thesespecies through selection and breeding.

Since its last Biennial Report, the SLP has extended its support to several more projects.Among them is a promising multi-partner project in West Africa that is testing a newapproach to on-farm research that could prove widely applicable and lead to attractive costsavings. Another project with considerable potential for impact involves the use of geneticmarkers to improve the feed value of crop residues in pearl millet and other (geneticallysimilar) grain crops and grasses. Lastly, the SLP has launched a project to “go virtual”. Thisproject is harnessing the potential of Internet-based technology to deliver efficiency gainsin the planning and implementation of research and the exchange of its results.


viii

The SLP conducts thorough ex-ante assessments before embarking on new research projects.Two such assessments are currently under way, one on existing dual-purpose cowpea researchin West Africa and the other on a new proposal to conduct research on the use of maize asfood, feed and fertiliser in Eastern and Southern Africa. The work on cowpea has shownthe crop to have great potential for income generation in areas where the market for livestockfeeds and products is growing. Studies on maize have highlighted the urgent need to reversethe steady drain of nutrients away from maize fields that is associated with the low use ofcommercial fertilisers on this crop and the application to other, more valuable crops of themanure produced by feeding maize residues to livestock.

The SLP emphasizes support to projects with the potential for near-term and/or transregionalimpact. The potential for near-term impact is greatest in the more productive ecoregions,where a broader range of technological options is feasible, but the prospects for impact inthe drier and more difficult ecoregions, such as the Sahel, has recently improved.Transregional impact is more likely to occur through the transfer of research methodologiesand in strategic research (for example in genetic markers work) than in the finishedtechnologies developed through adaptive research. A new project to assess the potential fortransregional impact has been launched.

Raising funds remains the most serious challenge facing the SLP. In future, the programmewill increase its emphasis on achieving transregional benefits from research, continue toseek efficiency gains in the conduct of research and devote more attention to assessingimpact.

ix

Preface

Contents

Introduction 1

The SLP in Action 3Projects in Progress 3Recently Launched Projects 27Ex-ante Impact Assessments 45

Meeting our Priorities 52Transregional and Near-term Impact through Technology Transfer 52

The Challenges Ahead 57

Annexes 59Sources 59SLP-supported Projects and their Status 62Financial Statement 65Centre Addresses and Contacts 67

1

Acting Locally

Introduction

Established in 1995, the System-wide Livestock Programme(SLP) of the Consultative Group on InternationalAgricultural Research (CGIAR) is one of several programmesdesigned to increase the returns to investments in theCGIAR’s research by linking activities at its different centres.

Livestock research in the CGIAR system is primarily theresponsibility of the International Livestock ResearchInstitute (ILRI). In founding the SLP, the CGIAR recognizedthat many of its crop-oriented centres, particularly thoseworking on improving the quality or yield of crops whoseresidues are used as feed resources, could also make an impacton livestock productivity. Similarly, the work of the centresconducting research on the management of natural resourcescould benefit through the inclusion of a livestock perspective.And there were also gains to be made in the field of policyresearch—another area of work shared by several centres inthe system.


2

Responsibility for establishing the SLP was assigned to ILRIwhich, in 1995, invited the participation of the otherCGIAR centres. Ten centres including ILRI joined theprogramme (see map on inside front cover). Under ILRI’sleadership, the partners developed a strategic plan for theSLP, which was approved by the CGIAR’s TechnicalAdvisory Committee (TAC), and constituted the LivestockProgramme Group (LPG) to agree the SLP’s agenda andadvise on its implementation. The LPG consists of arepresentative from each participating centre. A programmeplan, also developed in 1995, was revised in 1999.

The SLP operates as a decentralized team, whose membersparticipate directly in the research conducted at eachlocation. It also supports research by raising and channellingfunds to multi-partner projects selected for their potentialfor widespread impact. Centres are encouraged to join withnational institutions in ecoregional consortia or other formsof partnership to prepare research proposals. These areevaluated against set criteria by the LPG, which then decideswhether or not to recommend funding.

3

Acting Locally

The SLP in ActionThe SLP reports to its stakeholders biennially. Its first report,for 1997-98, focused on the rationale for the SLP, theevolution of Phase I of its programme and the progress madein establishing three ecoregional research projects with astrong livestock-related component. This report bearswitness to the considerable progress since made in imple-menting these and a number of other projects.

We report first on the three ecoregional projects. In the Kenyanhighlands, the SLP is making key inputs to a multi-partnerproject whose aim is to promote the development of small-holder dairying by addressing feed-related and otherconstraints. These inputs, which meet needs not covered byother project partners, are designed to enhance impact in thenear term by transferring technology already developed whileat the same time addressing longer term sustainability issuesassociated with soil fertility and biodiversity. In Latin America,the SLP supports a formally constituted ecoregional consor-tium responsible for increasing milk and meat productionfrom dual-purpose cattle herds raised in two sensitiveecosystems—the Andean hillsides and the forest margins. Thisproject has successfully developed some promising new feedresources that are now being extended to farmers. And in thedrylands of the Sahel and West/Central Asia-North Africa,four international centres have joined with national partnersto tackle dry-season feed shortages by improving theconservation, management and utilization of browse species.

Projects in ProgressIncreasing milk production in the Eastern and Central Africanhighlands

The cool highlands of Eastern and Central Africa cover only23% of the region’s total land mass yet are home to 51% ofits population. Rising populations and incomes are


4

increasing the demand for food, especially livestock products.The vast majority of rural people remain poor, undermin-ing their ability to respond to rising demand by increasingproduction sustainably. At the same time, the natural re-sources that support production are coming under increasedpressure, with erosion and declining soil fertility leading tofalling crop yields.

Intensively managed dairy cattle are an integral componentof the zone’s mixed crop-livestock farming systems, butinadequate feed quantity and quality constrain milkproduction. This in turn contributes to depressed crop yields,as less manure is available and farmers’ incomes are lower,reducing their ability to buy fertiliser. Under these condi-tions, legumes provide a key way forward for smallholdersseeking to increase their incomes from dairying whilesustaining their grain crop yields and protecting the fragilenatural resource base of the highlands.

Over the past decade, a number of national and internationalinstitutions, including the International Centre for Researchin Agroforestry (ICRAF), have been at work in the EmbuDistrict of Kenya, a fertile and well watered highland areawith considerable potential for smallholder dairy produc-tion based on crossbred cows. Researchers have successfullyintroduced new tree and shrub legumes, includingCalliandra calothyrsus, a fast-growing shrub that fits wellinto existing production systems and can take the place ofexpensive concentrates in feed rations. On-farm evaluationhas shown that planting 500 Calliandra trees—the numberrequired to meet the supplementary protein needs of a dairycow—increases household income by around US$ 130 percow per year, starting in the second year after planting. Italso saves farmers the time and money spent gathering orbuying fodder off the farm. This technology alone raisesfarmers’ incomes by around 10%. If it were widely extendedin Kenya, the benefits to producers and consumers wouldamount to about US$ 137 million annually.

5

Acting Locally

The project partners and their strengths

• Kenya Agricultural Research Institute (KARI):Plant breeding, production systems research, links with farmers, in-country networking

• Kenya Forestry Research Institute (KEFRI):Tree and shrub introduction and management

• National Dairy Development Project (NDDP):Milk production, feed resources

• National Agroforestry Research Project (NARP):Tree and shrub introduction, on-farm research, technology transfer

• International Livestock Research Institute (ILRI):Ruminant nutrition and smallholder production systems

• International Centre for Research in Agroforestry (ICRAF):Agroforestry, technology dissemination

• African Highlands Initiative (AHI):Regional networking and technology transfer

In 1996, ICRAF and its partners identified three contribu-tions the SLP could make to this work and submitted aproject proposal accordingly. The first is to enhance impactby identifying new areas to which technology can be trans-ferred and taking steps to promote transfer. This applieslocally, in Embu and other highland districts of Kenya, aswell as on a broader regional scale across Central and EasternAfrica. The second contribution is to support research toincrease the biodiversity of legume shrubs and other speciesavailable to farmers. This is desirable because the currentdependence on a handful of widely used species makes thesystem highly vulnerable to pests and diseases. For example,in 1992 Leucaena leucocephala, one of the most popularspecies tested at Embu, was heavily affected by an epidemicof psyllid, a leaf-eating insect from Southeast Asia. The third


6

contribution is to examine how smallholders manage anduse the relatively large amounts of manure produced by theircrossbred cows. Manure is one of the few organic sources ofnutrients widely available in the highlands, so the bestpossible use needs to be made of it. Research elsewheresuggests that much can be gained by combining the use ofmanure with small amounts of commercial fertilisers, whichremain too expensive for most farmers to apply atrecommended rates.

Most of the efforts in technology transfer have so far beendevoted to securing local impact. Household data fromprevious surveys in the central Kenyan highlands are beingintegrated with information in a geographical informationsystem (GIS) with a view to describing smallholderproduction systems, identifying broader recommendationdomains for the fodder species known to be of interest tofarmers, and developing strategies for promoting adoptionthat take into account local constraints and opportunities.By May 2000, preliminary recommendation zones had beenmapped for two popular legumes, Desmodium andCalliandra. In addition, the road infrastructure—animportant determinant of access to markets—had beenmapped in Embu and Meru. Field visits to ground truthsome of the data had also been carried out. Mapping wasdone in two phases: in the first phase, the recommendationdomain was identified using purely agro-ecological criteria,such as altitude and rainfall. In the second, the zone wasnarrowed by applying socio-economic factors such as accessto an all-weather road, the density of dairy cattle alreadypresent in an area, and the human population density(Figure 1). Next steps in the mapping exercise will includethe integration of data on soil type, marketing points andurban areas, and the extension of the exercise to westernKenya.

In parallel with this work, the researchers have stepped uptheir efforts on the participatory evaluation and dissemina-

7

Acting Locally

Combined recommodation domains

Natural domains

Socio-economic domainsbased on:•population density > 250pp/km2

•dairy density > 50 grade cows/km2

•distance to main road < 5km

Legend:

tion of leguminous fodder trees and other legume species.These activities involve a diverse set of partners, includingthe extension services, church and womens’ organizationsand schools, in addition to farmers and their associations.A group approach emphasizing farmer empowerment hasbeen adopted to maximize coverage and impact. The resultsare impressive: farmers’ groups have started some 250nurseries raising and distributing young Calliandra plantsto about 2600 farmers, 60% of them women. Training innursery establishment has been provided to the staff of threedepartments of the Ministry of Agriculture, two provincialadministrations and 13 non-government organizations(NGOs), besides numerous farmers. Farmers have also beenoffered training in the use of fodder technologies. Lastly,assistance has been provided to national and ICRAF staffdisseminating Calliandra and monitoring its uptake insouthwest Uganda.

Figure 1. Areas of Kenya with natural and socio-economic potentialfor Calliandra

Source: ICRAF et al (2000).


8

Despite the successes, several problems became apparentduring this dissemination work. Severe, prolonged droughtmeant that many nurseries were chronically short of water.And, mainly because of the drought, Calliandra seedlingsproved highly susceptible to pests such as crickets, hoppersand cutworms. The combined losses from drought and pestswere estimated at up to 50% of seedlings, halving impactand the returns to investments at a stroke. These lossesunderscore the urgent need to diversify the legume speciesand provenances available to farmers.

Further potential for spreading the use of Calliandra wasdemonstrated in an experiment on feeding goats, conductedby national scientists. Dual-purpose goats fed on Calliandraprovided more and better quality milk for family consump-tion, suggesting a powerful route for improving householdhealth and nutrition in poorer households unable to investin cattle. Children, who require vitamins and minerals inaddition to protein, stand to benefit particularly. Over 95%of farmers participating in this trial said they would con-tinue using Calliandra after the trial was over.

Research on diversifying the forage species available focuseson psyllid resistance in Leucaena, the use of low-cost locallyavailable alternatives to Calliandra and the potential ofherbaceous legumes.

9

Acting Locally

Some 15 psyllid-resistant Leucaena species and provenanceshave been identified, five of which are now being tested onfarm and on station. Most farmers identified L. tricandra asthe best biomass producer, followed by L. esculentapaniculata. One farmer was so impressed by L. tricandrathat she asked for seeds in order to raise her own seedlings.Data from the on-farm trials confirmed the farmers’ im-pressions, with L. tricandra showing the highest leafy biomassin all the harvests. Three other species—L. diversifoliadiversifolia, L. esculenta esculenta and L. esculentapaniculata—had similar, moderate levels of leafy biomass,while L. collinsii zacapana had the lowest yields, largelybecause it proved susceptible to psyllids after all.

The mulberry tree (Morus alba) is a locally available yetunderused tree species in traditional farming systems of thehighlands. It can be established at low cost from cuttings,and niches for expanding its use appear to exist on farms.The national researchers investigating this species plan toexpand their trials to cover other underused indigenousshrub species.

Several national research stations and projects have experi-mented with herbaceous legumes in Embu and otherhighland districts in the past. Their efforts have met withvarying degrees of success. Researchers at the Embu stationindicated some potential for a number of species that couldbe combined with napier grass, but the unavailability ofseed, together with establishment and managementproblems, have limited adoption. Agronomic studies onMucuna pruriens (velvet bean) showed considerable poten-tial for this legume, but no feeding studies on animals haveyet been undertaken. Stations such as Katumani andMtwapap have reported variable performance in Stylosanthesspp., Clitoria ternata and other species. Against this back-ground, it was thought appropriate to synthesize and sharethe state of existing knowledge before embarking on furtherparticipatory research with farmers. For this reason, a


10

workshop to which all interested scientists were invited wasscheduled for mid-June 2000.

While manure is commonly used by farmers to improvesoil fertility, many questions surround its use, which hasbeen under-researched in the past. Is the manure producedby the typical small dairy farm sufficient to meet the short-fall of nutrients required for cropping? Are cattle housingconditions conducive to the efficient collection and recyclingof manure? Can manure be improved by adding otherorganic materials, such as leaf prunings? Are inorganicfertilisers used in conjunction with manure? And to whatcrops is manure applied and how—by broadcasting beforethe season or by placing strategically during it? Research onthese topics aims to identify the key points at which inter-ventions could lead to greater efficiency.

As part of a broader effort to study farm nutrient balances,a survey on manure use was conducted on 26 householdsin the three major agro-ecological zones of the highlands inwhich dairy cattle are raised—the tea, coffee and maizezones. The survey distinguished aggregated farms, havingall their land around the homestead, from disaggregatedfarms with one or more plots away from the homestead.The reason for the distinction is that, owing to the difficul-ties of transporting manure, land holding patterns may bethe most significant factor affecting manure application. Thesurvey also divided farms according to whether or not theyhad cattle—the main livestock species producing manure.In addition, farmers were ranked for their wealth (1 = verypoor, 4 = wealthy) and asked whether or not they usedinorganic fertilisers.

The survey data are still being analysed, but initial resultssuggest that disaggregated holdings are a major disincentiveto manure use. Whether or not they raised cattle, 85-90%of farmers with aggregated holdings applied manure,compared to only 58-68% with disaggregated holdings.

11

Acting Locally

Surprisingly, whether or not a farm kept cattle had littleeffect on manure use, since farmers barter manure with eachother. Up to 90% of aggregated farms without cattle appliedmanure. Few farmers (generally under 22%) bought manure.Most of the farmers using manure were in wealth category2 (poor), which was the category to which the majority offarmers belonged. This suggests little manure is used by thevery poor, few of whom own cattle. It has, however, beenhypothesized that small farms with cattle have more fertilesoil than larger farms or farms without cattle. Thishypothesis, often found to be the case elsewhere, will betested in the next phase of research, in which animal housingconditions, alternative uses of manure (e.g. for vegetableplots) and the effectiveness of applications on crops such asmaize and napier grass will also be examined.

Improving dual-purpose cattle production in tropical LatinAmerica

Dual-purpose cattle, kept for milk and meat, account forabout 80% of national herds raised in tropical Latin America,but only 40% of the milk produced. Demand for meat andmilk is rising rapidly, as incomes and human populationsrise. Two factors hold back production: the low quantityand poor quality of feed available, especially during the dryseason; and the low genetic potential of the animals.

In 1996 a consortium of institutions led by the CentroInternacional de Agricultura Tropical (CIAT) launched aproject known as Tropileche, designed to increase milk andmeat production on smallholder farms by tackling the feedconstraint. Study sites were located in two ecologies wherethe needs were seen to be greatest, namely the relatively dryhillsides of Central America and the wetter forest marginsof the Amazon basin. Tropileche, which operates under theSLP, draws on CIAT’s expertise in forage development andland use and on ILRI for specialist inputs on the livestock


12

component. A wide range of research and developmentorganizations participate in the project at national level.Originally, sites were selected in central eastern Peru and inthe Central Pacific region of Costa Rica. More recently,research has been expanded to include Nicaragua, Hondurasand Colombia.

The approach adopted by the project combines on-stationresearch on the potential of new forage-based feed resourceswith participatory on-farm evaluation. Special emphasis isplaced on herbaceous and shrub legumes, with a view tointensifying mixed crop-livestock production at the sametime as protecting or enhancing the natural resource base.This component research is supported through diagnosticstudies to characterize the sites, economic analysis ofinterventions and surveys to determine their acceptabilityand adoption.

In the hillside areas of Costa Rica, on-farm research onCratylia argentea has shown this protein-rich shrub legumeto be an effective supplement to replace the more expensiveprotein concentrates used during the dry season. Milk yieldsobtained when feeding the shrub in various combinationswith grasses were similar to those obtained using concen-


• National research institutes, universities, NGOs and farmers’ associations (Peru, Costa Rica,Nicaragua, Honduras and Colombia):

Dairy development, production systems research, links with farmers, in-countryand regional networking

• Centro Internacional de Agricultura Tropical (CIAT):Forage development, land use

• International Livestock Research Institute (ILRI):Ruminant nutrition, livestock production systems

13

Acting Locally

trates. Combined with sugarcane, fresh Cratylia can increasemilk yields by 1-2 litres per day (20-30%). The shrub canbe grazed in the field or fed in a cut-and-carry system. Cut-and-carry systems require some extra labour and, whensugarcane is included in the diet, investment in a shredder.A further advantage of Cratylia is that it can be turned intosilage during the rainy season, when it is not needed, andfed later on, taking the pressure off overgrazed pasturesduring the dry season. Indeed, this technology may allowsome land to be freed up for other uses, such as reforesta-tion (Figure 2). Again, feeding Cratylia as silage is moreeconomical than using purchased commercial concentrates.It is thought that, if well managed, a stand of Cratylia canbe grazed or cut for up to 20 years. The shrub is well adaptedto a range of soils, including those with low fertility, andcan withstand severe drought. It needs a small amount ofphosphorus to get started, but this can be supplied usingcow manure from the milking shed.

Farmers who have tested Cratylia are enthusiastic about it.A farmer in the Central Pacific region who grew 1 ha of thecrop on degraded land said that it had increased his netincome by about one-third. On five farms, there is evidenceof soil improvement derived from this legume’s ability tofix atmospheric nitrogen. Farmers beyond the research sitehave started adopting Cratylia spontaneously, and commer-cial seed producers are multiplying and marketing seed.

Another forage being tested by Tropileche is the legumeArachis pintoi, a wild groundnut plant native to Brazil thatis suitable for use in both the hillsides and the forest margins.This is planted directly in pastures already improved throughintroduction of the African grass, Brachiaria. This formida-ble combination of plants provides thick ground cover,preventing erosion and enhancing fertility even on poor acidsoils, in addition to a high-quality feed that greatly increasesmilk yields. A 30% legume content in the sward is sufficient


14

to increase milk production by 8% over straight grass plussupplementation with 1 kg of concentrates per day. It alsohelps increase biomass yield in pastures over time—a valu-able contribution in areas where they are renowned for theirtendency to degrade.

Also being tested in the forest margins is Stylosanthesguianensis. This herbaceous legume has been widely testedin tropical America, but is still underused. However, it couldbecome more popular in the future, as more farmers startto specialize in milk production for market. Research inColombia and Peru has shown that feeding Stylosanthes topre-weaned calves increases the amount of milk for sale by21% and boosts liveweight gain by up to 30%. In addition,growing Stylosanthes in rotation with rice increased the yieldof the following rice crop. The use of Stylosanthes could thuseliminate the need to fallow land in a shifting cultivationsystem, helping to reduce deforestation. The critical ingre-dient in the system is farmer management, since farmersmust resist the temptation to overgraze Stylosanthes pasturesif these are to persist through the 2 to 3 years needed torestore soil fertility before the next rice crop can be grown.Growing Stylosanthes is less expensive than establishing a

Figure 2. Percentage of pastureland that could be allocated toother uses if improved foragesare adopted

50

40

30

20

10

0

(%)

Cratylia +sugarcane

Brachiaria +Cratylia +sugarcane

Brachiaria +Arachis +Cratylia +sugarcane

27.5 36.50

Source: Holmann et al (1998).

15

Acting Locally

legume-grass pasture and results in higher cash flow frommilk without affecting calf growth.

At Pucallpa, in Peru, farmers seem unlikely to adopt any ofthe feed technologies being tested at present. Cattle numbersfell sharply due to depredation during the long guerrillawar in the countryside. A low human population densityand long distances to markets are further factors reducingthe potential for improved dairy production at this site.Consequently, Tropileche’s partners have decided to investtheir efforts in a new site, Moyobama, also located in thePeruvian Amazon but closer to a major market. Here farmersare improving the genetic potential of their herds for milkproduction and a cooperative milk processing plant has justopened. There is, therefore, a demand for improved feedingsystems. Work continues in the forest margins of Colombia,where the technologies being tested do appear viable.


16

The research of Tropileche has demonstrated that legume-based technology significantly improves milk production,is a viable substitute for expensive feed concentrates and isenvironmentally friendly. Small-scale farmers can easilyadopt and manage such technology, even when they lackaccess to credit. Access to market seems to be the main factoraffecting farmers’ willingness to adopt.

Besides technology development, Tropileche also has amandate for technology transfer. The consortium uses severalmethods to disseminate its results quickly to new areas.Conventional methods such as field days are still an effec-tive way of eliciting interest on the part of farmers,extensionists and NGO workers. Modern, media-basedmethods are also being used. In 1999, an 11-minute videowas produced featuring the experiences of a small CostaRican farmer who has adopted Cratylia- and Arachis-basedtechnologies. This farmer has doubled his family income,now produces more milk from less land and uses some ofthe freed up area to produce timber and protect waterresources. The video, which is in Spanish, will be widelyshown to farmers in both Costa Rica and elsewhere.Tropileche also publishes a biannual newsletter for forageresearchers throughout the region, together with a journal,Pasturas Tropicales, which has over 500 subscribers. Lastly, awebsite, http:/www.ciat.cgiar.org/tropileche/start.htm,contains the latest research results in addition to some 2200other references, provides a list of relevant researchers in theregion and allows visitors to interact with one another.

One measure of Tropileche’s impact is that new countriesare now joining the research effort, using bilateral funds. InNicaragua and Honduras, local partners have establishedover 65 ha of improved forages on 20 farms at five differentsites. Ecuador and Bolivia have also expressed interest andpartners there are currently preparing research proposals.

In October 1999, CIAT held a regional consultation toestablish priorities for the dairy sector in Central America

17

Acting Locally

and the Caribbean and hence for Tropileche’s activities inthese subregions. Information for the workshop was gath-ered independently in Costa Rica, the Dominican Republic,Honduras, Jamaica, Nicaragua and Panama. Reports fromthese countries served as the basis for discussion at the work-shop, which was held in Atenas, Costa Rica.

The main challenge facing Tropileche is to scale up—toextrapolate the successes achieved in small areas to muchlarger, similar areas. The next phase of the project will there-fore place more emphasis on technology dissemination ratherthan technology development. Seed supplies are a criticalingredient of successful scaling up. So far, 20 seed bankshave been established, but more farmers and companies witha capacity for seed production need to be identified. Anotherneed is accurate identification of areas with similar soils,climate and socio-economic factors to those in which a giventechnology has already proved successful. GIS resources atCIAT are being harnessed to meet this need.

Improving dry-season feed resources in the semi-arid rangelandsof the Sahel and West/Central Asia-North Africa

The vast semi-arid rangelands of the Sahel and of West/Central Asia-North Africa are home to millions of cattleand small ruminants, raised largely by poor pastoralists andagropastoralists. As cropping and other forms of land useexpand into these lands, livestock are increasingly confinedto the less productive areas, where they face severe shortagesof feed, especially during the long dry season. In 1996, theInternational Center for Agricultural Research in the DryAreas (ICARDA) joined forces with three other interna-tional centres and with national institutes in nine countriesin a new bid to overcome the feed constraint by developingthe use of multi-purpose fodder shrubs and trees. The SLPagreed to fund the proposal and work began in November1997.


18

Many past attempts have been made to increase the feedsupplies available to livestock in these dry lands. Theseattempts have been for the most part fragmented, with poorcommunication between researchers. They have also tendedto be non-participatory, with solutions developed byscientists being introduced to pastoralists and farmerswithout their effective involvement. As a result, fewtechnologies have been widely adopted and animalproductivity remains low. In the Sahel, the poor market forlivestock produce has been a further disincentive to adoption.But in recent years incomes in the Sahel have started to riseand the concentration of people in cities has begun toconstitute a significant unmet demand. Peri-urban produc-tion systems are, consequently, becoming increasinglymarket-oriented.

Now is therefore the time for a new attempt to develop anddisseminate productivity-increasing technology. The SLPcan help by ensuring a bottom-up approach to project


• National institutes (Burkina Faso, Jordan, Mali, Morocco, Niger, Pakistan, Senegal, Syriaand Tunisia):

Production systems research, links with farmers and NGOs, in-country and regional networking

• International Center for Agricultural Research in the Dry Areas (ICARDA):Dryland ecosystems in West/Central Asia-North Africa, barley, sheep

• International Centre for Research in Agroforestry (ICRAF):Agroforestry, multiple uses of trees

• International Crops Research Institute for the Semi-Arid Tropics (ICRISAT):Dryland ecosystems in the Sahel, pearl millet, sorghum

• International Livestock Research Institute (ILRI):Browse fodder, ruminant nutrition

19

Acting Locally

planning, in which proposed research is made to reflectproducers’ real needs. It can also ensure a participatoryapproach to testing new technologies with farmers. And itcan link projects, ensuring that knowledge and results aretransferred to similar locations.

Temperature conditions across the semi-arid rangelands varygreatly, but the northern swathe of rangelands in West/Central Asia-North Africa experiences cold winters, withup to 35 days of frost each year. Syria, Uzbekistan andPakistan are collaborating with ICARDA in screening arange of shrubs for their tolerance to cold, includingArtemesia spp. and Salsola spp.

Even if shrubs are tolerant to cold, establishing them in thedry rangelands is difficult. Many sites are now almost bareof vegetation, making for very harsh, exposed conditions.Surface soil is often crusted and a black moss that spreadsrapidly over the soil surface also thwarts emergence.

Working with the Syrian Steppe Directorate, scientists atICARDA are testing different establishment methods. Atpresent, the Directorate produces shrub seedlings in a nurs-ery, transports them and plants them in areas prepared bydeep ripping, which is expensive. Instead, the scientists aretesting direct seeding, focusing on low-cost, low-labourmethods acceptable to local pastoral communities andauthorities. Three methods are under test: traditionalbroadcasting on the soil surface with no soil preparation;harrowing before seeding, to break the crusted soil surfaceso as to improve emergence; and the digging of shallow pitsto facilitate run-in and promote the accumulation of organicmatter. To dig the pits, a special pitting machine is used,consisting of a simple modified disc plough mounted on atwo-wheeled trailer pulled by an ordinary vehicle. The resultsobtained so far suggest that pitting is a more efficient estab-lishment method than broadcasting or harrowing, at leastfor Salsola shrubs.


20

One of the hardiest shrubs for these Mediterranean areas isthe indigenous saltbush (Atriplex halimus), which variesgreatly in its palatability. ICARDA and its partners are test-ing different provenances for their effects on sheep, the majorlivestock species raised in the region. The shrub is also be-ing tested for its suitability as an intercrop with barley, theregion’s major feed as grain, straw and stubble. The results

so far show that barley yields are higher when intercroppedwith Atriplex than when monocropped. Average grain yieldwas 443 kg/ha in intercrop plots, compared with 360 kg/hawhen barley alone was grown. Stubble yields were also mar-ginally higher in intercropped plots. Animals grazing theplots ate similar amounts of stubble on both monocroppedand intercropped plots, but on the intercropped plots theyalso consumed an additional 0.15 kg/day of saltbush foliage,providing them with a valuable additional source of proteinto complement the energy obtained from the barley. Theyalso spent longer grazing the intercropped plots. These resultssuggest that intercropping barley with Atriplex could greatlyincrease crop and animal production at the same time ashelping to protect fragile soils from wind and water erosion.

In Morocco, encouraging results were achieved in on-stationand on-farm testing of Atriplex in combination with a rangeof grain crops, including barley, oats, a barley-fodder pea

21

Acting Locally

mixture and an oats-vetch mixture. Total biomass and grainyields were higher in intercropping systems, as also wereenergy and crude protein yields, which increased by 11-93% and by 16-196% respectively. Intercropping improvedland-equivalent ratios to 1.20-1.46, suggesting that this tech-nology will be particularly useful in areas where farm size issmall. In each of two rural communes, farmers have nowplanted some 200 ha of shrubs in combination with grainand forage crops.

In Jordan, scientists are studying the effect of plant densityon the survival and productivity of Atriplex halimus and A.nummularia. Survival rates were lower at higher plant den-sities, reflecting water scarcity and heat stress in summer,even when water harvesting is used. A. halimus had a highersurvival rate than A. nummularia.

None of the shrubs available to livestock in the semi-aridlands can by itself provide a completely satisfactory diet tolivestock. Research in Tunisia focuses on the potential ofcombining three common shrub species, Acacia, Atriplexand Opuntia, to give a better balanced diet. Acacia has arelatively high crude protein content but also containstannins that reduce its feed value by forming insoluble com-plexes with proteins. The addition of a chemical,polyethylene glycol (PEG), to the diet improves perform-ance by breaking down the complexes, but its use is limitedby its high cost. Increased energy and nitrogen in the dietmight perform the same service more cheaply and efficiently.Opuntia ficus, or spineless cactus, is one of the commonestplants found on rangeland and is a valuable source of water,ash and some vitamins, besides having a high energy value.Atriplex, in contrast, has a low energy value but relativelyhigh nitrogen. Its high salt content means that it needs tobe eaten with copious amounts of water, to ensure that theanimals excrete the salt. This makes it the perfect partnerfor Opuntia. Together, the three shrubs should in theoryprovide a healthy diet.


22

The preliminary results of this research confirmed thishypothesis. A diet in which Acacia cyanophylla was supple-mented by barley maintained the weight of sheep, whereassupplementing with Opuntia alone led to weight loss. Adiet of Atriplex and barley straw alone lacked sufficientenergy, also resulting in weight loss. Adding barley grain orOpuntia to this diet resulted in weight gain, with Opuntiaproving somewhat more effective than barley.

South of the Sahara, conditions are hotter than in the Medi-terranean zone, but pressures on the resource base are equallysevere. Levels of poverty are, if anything, worse, making theneed for low-cost improvements in feed availability andquality all the more necessary. High-quality feed is espe-cially scarce during the long dry season, when grasses witherand die. Browse trees and shrubs whose foliage remains suc-culent and rich in protein during this period are especiallyvaluable.

In Burkina Faso, a team of researchers is building on theagroforestry research of the 1980s to pursue increased useof local trees and shrubs. In 1997, the team conductedsurveys to determine the priority species meriting conser-vation, increased utilization and, in the longer term, geneticimprovement. The survey covered the country’s three majoragro-ecological zones, including the northern Sahelian zone,

23

Acting Locally

with annual rainfall lower than 600 mm, the northernSudanian zone, with rainfall between 600 and 850 mm,and the southern Sudanian zone, with rainfall of 900 to1300 mm. The presence and diversity of trees in the land-scape increases as rainfall rises, with species such as Vitellariaparadoxa becoming increasingly common. The surveycovered 20 villages in the northern Sudanian zone, whereFaidherbia albida, Parkia biglobosa and V. paradoxa are found.A further four villages in the Sahelian zone were covered,where thorny species and Hyphaene thebaica predominate.Lastly, a sample of three villages in the southern Sudanianzone were included. Here, Borassus aethiopium, V. paradoxaand P. biglobosa are found. A multidisciplinary team askedwomen, working adult males and old people to list 15important agroforestry species and then to select their 10priority species. In each village, interviewees were also askedto name the species used for eight different purposes—food,feed, medicine, soil fertility, fuelwood, building, crafts andany others—and to say how they would like to see theirpriority species improved. Lastly, the team evaluated thetraditional knowledge of farmers as a means of definingselection criteria and procedures.

Farmers in the three zones named over 70 agroforestry speciesas useful. The top 10 species differed in each zone, butV. paradoxa and P. biglobosa both featured high on the list inthe central and southern zones. Listings varied consider-ably between different social groups and places. For example,women in the southwest of the country almost invariablyincluded palmyra (B. aethiopium) among their first sixspecies. The nearby town of Banfora is a centre for craftitems made from this species by women, for whom this is amajor source of income. Around Korea, in the northernSahelian zone, in contrast, F. albida, H. thebaica andA. digitata were the first three priority species. The first twoof these are vital sources of income, with Faidherbia podsconstituting a protein-rich supplement for fattening sheep


24

while Hyphaene (the so-called doom-palm) is used forweaving and for its popular edible fruits. Income-earningpotential was often the most important criteria affectingthe choices made by different groups. Gender roles withinthe household and male or female usufruct or ownership ofdifferent tree species also strongly affected choices. Forexample, women around Wolonkoto and Siniama in thesouthern Sudanian zone rarely named the doom-palmamong their priorities because this species is owned andmanaged almost entirely by men. The use of the species islittle developed here.

Forage species were found to be valued most in the centraland eastern areas of Burkina Faso’s Central Plateau, wherethe highest concentrations of livestock are found. Livestockproducers recognize intraspecific variability in forage qual-ity, which is reflected in such characteristics as palatability,leaf or grain size, age of tree and amount of foliage produced.Several important forage species have multiple uses, includ-ing Ziziphus mauritania, Balanites aegyptica and Adansoniadigitata. Despite their importance, forage species are rarelyplanted by farmers, mainly because seedlings are not availableor because of the difficulties of protecting young trees.Farmers also said they had no knowledge of how to raiseseedlings and plant trees.

The study has revealed that the genetic heritage of thesemultipurpose species is under threat. The next step will beto assess the heritability of key traits in each species, so thatin situ genebanks can be established and a breedingprogramme launched. At the same time, sustainability-oriented management practices need to be developed anddisseminated, using a participatory approach.

Research in Niger focuses on the grazing and browsingbehaviour of different livestock species, with a view toidentifying the priorities for conserving and improvingbrowse species and managing rangeland. A study conducted

25

Acting Locally

during the dry season in a village west of Niamey, in theSahelian zone, showed that the relatively unpalatable Guierasenegalensis now constitutes nearly 99% of available browsebiomass. More palatable species, such as Boscia senegalensis,B. angustifolia and Piliostigma reticulatum, are scarce. Goatsdevote 30% of their grazing time to eating B. aegyptica andAcacia spp., which represent less than 1% of all availablebrowse. Species such as these urgently need conserving.

Also in Niger, research by ILRI and ICRAF has focused onCombretum aculeatum, a broadly distributed species muchappreciated for its nutritive value and palatability to sheep.Little is known at present about the factors influencing itsforage quality and the potential for selecting superiorprovenances or genotypes. As part of a broader germplasmcollection programme run by ICRAF, seeds of 60 trees werecollected in 1993 from a 20-km wide area along the NigerRiver. Half-sib families were produced from the seeds bygrowing them in a nursery. In 1994, foliage from the newgeneration of trees was analysed for its nutritive value. Mosttrees had relatively high concentrations of protein and phos-phorus, confirming the importance of this species as a sourceof these nutrients. The results supported the hypothesis thattree genotypes with superior fodder quality can be identified.

In Senegal, national researchers have investigated a range ofspecies for their feed value and for various management


26

parameters, including time and methods of harvesting andpost-harvest drying and storage. Palatability, nutritive valueand drought resistance are the three most important criteriadetermining the species on which research should concen-trate. Acacia albida, G. senegalensis, F. albida, A. digitaria,Pitecellobium dulce and Calotropis procera all fulfil theserequirements, but their foliage also contains some anti-nutritional elements, including tannins. The results ofharvesting trials showed that moderate harvesting of foliagemakes the tree more likely to survive and to recover quickly.For all the species studied, the best time of year to harvest isthe early dry season, when forage is at its most nutritive andtannin levels are not yet dangerous to animals (Figure 3).However, the availability of nutritious grasses and the generalabundance of feed at this time mean that browse foliageneeds to be preserved for use later in the dry season, whenfeed is much scarcer. The scientists tested several preserva-tion methods, including drying in the sun or in the shade,oven drying (at 60oC and 80oC) and ensiling. Oven drying

Figure 3. Seasonal variations in tannin content of browse species in Senegal

Late wet Early dry Late dry Early wet

Tannin content(g gallic acid as % of DM)

6

5

4

3

2

1

0

A. albidaA. digitataC. proceraL. leucocephalaP. dulce

Season Source: Fall et al, in Gintzburger et al (2000).

27

Acting Locally

and ensiling gave temperatures that are too high, reducingfeed value. Sun or shade drying made little difference tofeed value and can be recommended as the simplest andcheapest methods, leading to feed values that remainrelatively stable throughout the dry season. Feeding trialsconfirmed the presence of anti-nutritional elements,indicating that the proportion of browse in diets should belimited to 15% for the foliage of G. senegalensis and thefruit of F. albida and to 30% for the foliage of A. digitata.Foliage of P. dulce can safely be fed in amounts up to 50%of the ration, while with C. procera the level can be raised to75%.

In line with the SLP’s mission to ensure the transfer ofresearch results, a workshop on the project’s findings so farwas held in February 1999. All the national SLP collabora-tors from participating countries attended and presentedtheir results, which are summarized in a workshopproceedings (Gintzburger et al, 2000).

Recently Launched ProjectsSeveral projects that were at the planning stage last time theSLP reported are now being implemented. We have selectedthree on which to report this year. The first is a particularlypromising project on improving crop-livestock productionsystems in the dry savannas of West Africa. This project hasdeveloped a new approach to multi-partner on-farm researchfrom which considerable synergistic effects and spilloverbenefits can be expected, generating useful cost savings.While this project operates by concentrating resources, thepartners in our second new project are widely dispersed geo-graphically. They are united by their use of genetic markersto improve the feed value of a number of widely grown,genetically related, grain crops. This project contributes to,and builds on, successful research already undertaken byILRI and ICRISAT on pearl millet and sorghum, two of


28

the world’s most widely grown dual-purpose grain crops.Again, considerable spillover benefits can be expected.Global multi-partner research of this kind has high transac-tion costs when conducted by conventional means. Ourthird project, in which Internet-based technology is beingused to create a “virtual SLP”, is laying the foundations forsizeable gains in the efficiency with which widely dispersedpartners will be able to collaborate.

Improving crop-livestock systems in the dry savannas of WestAfrica

Dry savannas, with annual rainfall between 400 and 900mm and a growing period of 90 to 150 days a year, covermore than 50% of the land area of sub-Saharan Africa. Soilsare poor, especially in nitrogen and phosphorus. Humanand animal populations are high and increasing, driving theintensification of both crop and livestock production. Fallowperiods are disappearing and cropping is expanding intomarginal areas where rainfall is too unreliable to allow theuse of commercial inputs. Such trends could, if unchecked,have disastrous consequences for the food security of poor,small-scale farming families and for the environments inwhich they live.

Researchers have identified three major crop-livestockproduction systems found in the savannas (Figure 4). Inthe drier, more northerly areas, pearl millet and cowpeadominate the cropping system, livestock husbandry is largelyextensive and transhumance is practised. Settled farming is,however, on the increase. In the wetter areas to the south,sorghum, cowpea and groundnut are grown, often withmaize, soybean and cassava as secondary crops. Livestockhusbandry is more intensive and there are many moresettled crop-livestock farmers. This system overlaps with athird system, in which cotton and groundnut are impor-tant cash crops and farmers tend to use somewhat moreinputs.

29

Acting Locally

Figure 4. Major crop-livestock production systems of the West and Central African savannas

Source: Grandi (1996).

Research on these systems has been pursued in northernNigeria for many years. All involved have come to realizethat farmers’ needs can only be met if the componentresearch conducted by different individuals and institutionsis integrated in a holistic approach. The SLP is supportingthe process of integration by contributing funds to a multi-partner project working at different sites across the region.

The evolution of the project reflects researchers’ growingrealization of the need for integration. The InternationalInstitute of Tropical Agriculture (IITA) has long conductedresearch on cowpea, but until recently this research focusedon developing high-yielding grain varieties for cultivationin sole cowpea fields, using pesticides to protect the crop. It

Annual rainfall(mm)

Climatic zone Livestock system

Crop system

150

600

1200

Sahara

Sahel

Sudan

Nomadism

Transhumance

Dry-seasonpastures

Settled

Rainfed Irrigated

Limit of cropping

MilletCowpea

RiceVegetablesSorghum

MaizeGroundnutCotton

Systems 1, 2 and 3*

* Note that systems 2 and 3 overlap


30

is now recognized that cowpea is rarely grown in isolationby farmers but is almost always intercropped with sorghumor pearl millet, that most farmers can barely afford pesti-cides, and that farmers in any case value the residue of thecrop for use as forage just as much as the grain for humanconsumption. By the mid-1990s, scientists at ILRI had be-gun working with IITA breeders to investigate the foddervalue of different cowpea varieties, while scientists fromICRISAT provided inputs to the cereal components of thefarming system. In 1997, scientists from the three institutesmet and agreed that, in the following year, they would pooltheir resources in order to develop a more holistic approach


• National research institutes/extension services:Crop improvement, crop residue and manure managementDevelopment of fodder resourcesLinks with farming communities

• International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)Pearl millet, sorghum, groundnutCropping systems, farmer participatory research, modeling

• International Institute of Tropical Agriculture (IITA)Cowpea, maize, cassava, soybeanSocio-economics and natural resource management

• International Livestock Research Institute (ILRI)Livestock production, income generation, nutrient cycling, crop residue managementLivestock-related policy and institutional issues

• International Fertilizer Development Center (IFDC)Integrated nutrient managementInput-related policy and institutional issues

• Centre for Overseas Research and Development (CORD), University of Durham, UKParticipatory research on nutrient cycling and natural resource managementSocio-economic and rural livelihoods issues

31

Acting Locally

to on-farm research. The scientists met again in early 1998to plan their research for the coming season and, inrecognition of the importance of soil fertility, were joinedthis time by scientists from the International FertilizerDevelopment Center (IFDC). It was at this meeting thatthe best-bet approach was developed (see Box).

Because this new approach was challenging, it was decidedto start small. A pilot study was therefore launched in Bichi,a village north of Kano in northern Nigeria. After the firstseason, in November 1998, the SLP provided funds to holda workshop to present the strategy used, review progress inthe pilot project and develop an approach for a larger regionalproject. Representatives from national institutes across theregion were invited to attend, together with relevant inter-national scientists and donor agency representatives. The

Integration for impact: The best-bet approach

When the scientists at the meeting planned their research solely from the point of view of thecomponent for which each was responsible, the resulting number of treatments was too high to bemanaged by farmers. Each scientist had proposed testing several varieties of each crop, variouscrop management and crop geometry options, and a number of treatments and practices relatingto crop residue and manure management.

The scientists agreed to reduce the number of treatments by selecting and combining what theythought were the best options within each component. The result was the best-bet approach.

Two types of situation were recognized—one in which farmers might use small amounts of inputsand one in which they would not. The final number of treatments was therefore three: a best-betset of options without inputs (BB), a best-bet set of options with inputs (BB+) and the localfarmers’ traditional practices (L).

It was decided that the options would be evaluated with farmers rather than for them, and thatthe whole range of factors important to them would be covered in a single set of trials. Thesefactors included crop grain and residue yields, grain quality, post-harvest losses, the feeding oflivestock, manure applications and their impact on yields, and labour and other socio-economicimplications.


32

Centre for Overseas Research and Development (CORD)of the University of Durham, UK had by this time expressedits interest in the project and was also represented. A re-gional project proposal based on ideas and outputs fromthe workshop was submitted to the SLP in 1999. The SLPhas already provided funds to expand the work at Bichi,include a second site in northern Nigeria and start activitiesin Niger and Mali.

The results achieved at Bichi confirm that this newapproach has exciting potential. In the 1998 season, dry-matter yields of sorghum fodder, cowpea grain and cowpeafodder were all highest in the BB+ treatment. The fodderyield of improved sorghum did not differ significantly fromthat of local sorghum, but this was mainly because the localsorghum is very tall and has thick stems. It was estimatedthat only about 30% of the dry matter of the local sorghumwas edible, compared with 60% in the improved sorghum.The stalks of local varieties are, however, valued by farmersfor use as building and thatching materials. The mostdramatic difference in yields occurred in cowpea, the grainyield of which was 16 and 2 times higher in the BB+ treat-ment than in the L and BB treatments respectively. Cowpeafodder yields in the BB+ treatment were also higher, by afactor of 1.5 over the BB treatment and 5 over the Ltreatment. Some losses in both sorghum and cowpea fodderyields occurred during transportation and storage, suggest-ing opportunities for improvements in these areas.

As part of the BB and BB+ treatments, the scientists hadoriginally suggested that the farmers plant a second crop ofcowpea after the first. This is made possible by the shorterduration of the improved cowpea varieties developed byIITA. Previous research had shown that, provided the firstcrop is harvested by the end of September, it is possible toobtain a good fodder yield and some grain from the secondcrop. However, farmers at Bichi were ambivalent about thisoption, for two reasons. First, sowing the second crop

33

Acting Locally

coincided with the peak family labour requirement forharvesting tomato and pepper and, second, in 1998 the rainscontinued unusually late into October, so that the firstcowpea crop continued to produce pods and farmers wereable to go on harvesting grain. Nevertheless, a few farmersdid plant a second crop and obtained useful amounts offodder and grain.

In the 1999 season, yields of both the grain and the legumecomponent were generally somewhat lower, reflecting lowerbut more typical rainfall. At the farmers’ request, the BBtreatment was altered to allow the use of inputs, but withlocal instead of improved varieties. This produced an inter-esting result: the local sorghum variety yielded more grainthan the improved one. In both the BB and the BB+ treat-ments, grain and fodder yields were twice as high and 1.5times as high as in the L treatment.

Research at the second site in northern Nigeria, UnguwanZangi, began in 1999. This site is wetter than Bichi, leadingto the testing of different cowpea and sorghum varieties.Here the BB and BB+ treatments both gave higher cowpeagrain and fodder yields than the L treatment. Farmers wereexcited by the potential of the new varieties, and expressedan interest in earlier planting and more timely spraying inthe next season, both of which should raise yields still higher.Sorghum grain and fodder yields were also significantlyhigher in the BB and BB+ treatments. All the farmers werepleased with the performance of the improved sorghumvariety and many said they would start using it elsewhereon their farms.

The site in Niger is considerably further north than those inNigeria, so pearl millet takes the place of sorghum as themajor food grain. Here the local cowpea variety did betterin terms of both grain and fodder yield than the two im-proved varieties tested. This site was also chosen for a trialon the effects of fertiliser on crop yields. Locally available


34

Tahoua rock phosphate and standard NPK compoundfertiliser were applied in small quantities (13 kg and 4 kgP/ha respectively), using pocket placement for the com-pound fertiliser. Yields rose dramatically at these low ratesof application, which are considered affordable by farmers.

Work in Mali is at an early stage, involving characterizationof sites in three regions typifying the country’s crop-livestocksystems. Soil profiles have been analysed and a GIS used tomap each of five villages in each region. A typology of farmsbased on the extent to which animal traction is used hasalso been completed.

The project’s experiences have revealed several topicsdeserving increased attention in the future. A significantfactor affecting profitability of the best-bet sets of options isthe storage of cowpea grain. If farmers sell grain immedi-ately it is harvested, it fetches roughly one-third as much asit does if stored and sold 3 months later. Stored cowpea is,however, susceptible to insect pests, especially weevils.Possibly, the best-bet sets should therefore include cowpeavarieties whose seeds have some tolerance to weevils. Asimple storage method involving layered polythene bagsmight also be an appropriate solution.

The sets need to be tailored to cover a range of situationswith regard to the use of inputs. Farmers with poor marketaccess may not be able or willing to use any inputs at all. Inother areas, however, the use of inputs is increasing as marketsdevelop. In such cases it will be necessary to select the cropvarieties included in the set carefully, so as to ensure thatthey yield better than local varieties whether or not inputsare used.

Declining soil fertility is a problem throughout the zone.Most farmers cannot afford large amounts of chemicalfertiliser. And farmers are often lucky if they can find anycommercial fertiliser on the local market at all. Under thesecircumstances, strategies and technologies that maximize the

35

Acting Locally

benefits from limited amounts of inputs are needed. Thismakes it all the more important to integrate the livestockand crop components of these systems, since livestock arean important source of manure in addition to tractive power.

Experiences in the first 2 years have constantly challengedresearchers’ assumptions. In Niger, for example, the “improved”crop varieties in the best-bet set of options were not in factsuperior to the farmers’ local varieties. This has led to thepossibility of a best-bet set consisting of farmers’ varietiesplus other innovations and has once again confrontedresearchers with the need to take fodder yields seriously whendeveloping new varieties. At Bichi, in contrast, farmersindicated a preference for the improved sorghum variety onaccount of its fodder quality. However, farmers here havebeen reluctant to adopt another innovation proposed byresearchers, the growing of a second cowpea crop. Somebest bets may be fairly location-specific, so it will be vital tospecify to which region or system they apply.

The new research paradigm being tested by the project iscontinuing to evolve. However, its potential for refining newtechnologies and promoting their uptake has already becomeapparent. Too often in the past, researchers have adopted a


36

multidisciplinary approach to system diagnosis, only torelapse into disciplinary component research for subsequentstages of the research process. The holistic approach adoptedin this project is helping to overcome this problem. That iswhy the SLP has developed a major proposal to increase thefunding available for testing and disseminating the approach,which promises to deliver the long-awaited impact on thisregion’s increasingly stressed mixed crop-livestock produc-tion systems. Impact here will have implications for otherregions of the developing world with complex and resource-poor farming systems, a consideration that furtherstrengthens the case for SLP support.

Enhancing the feed value of crop residues

As grazing resources decline owing to the expansion ofcropping and the disappearance of fallow periods, livestockbecome increasingly dependent on crop residues for theirfeed supplies. Scientists have attempted to improve the gen-erally poor nutritional value of these residues by chemicaland biological means, but the results have not been widelyadopted by resource-poor farmers, mainly because of theinvestment and labour implications. Genetic enhancementoffers a more promising way forward since the results canbe incorporated in improved seed, a technology moreacceptable to such farmers. The genetic route to improve-ment is also more environmentally friendly than chemicalor biological treatments. But despite these advantages,improving the feed value of crop residues has seldom been apriority for plant breeders, who have tended to focus almostentirely on grain production for human consumption.Modifying this emphasis is part of the SLP mandate.

Since the 1980s, ILRI has collaborated with ICRISAT toimprove the feed value of pearl millet and sorghum, twodual-purpose cereals vital to the survival and livelihoods ofresource-poor farmers across the semi-arid tropics. Early out-

37

Acting Locally

comes of this research, which was initially conducted mainlyin West Africa, were the discovery of a number of traitsindicative of feed value, including the stay-green trait insorghum and the brown mid-rib and trichome-less traits inpearl millet. These findings encouraged the two centres tocontinue their efforts through further research based atICRISAT’s headquarters in India. In this second phase ofresearch the number of partners involved was expanded toinclude specialized institutes in the UK, the USA andAustralia, in addition to relevant national programmes inIndia. Using an approach firmly grounded in an assessmentof farmers’ priorities, the researchers have applied the tech-niques of marker-assisted selection (MAS) to improve theprecision with which relevant traits can be incorporated ina more productive genetic background and to speed up thedevelopment of new dual-purpose varieties.

The SLP has become involved in this work because of thepotential for transferring its results to other plant speciesused all or in part as animal feed, thereby capturing signifi-cant spillover benefits from the investments already made.Pearl millet has a fairly simple genome, making it relativelyeasy to map and so shortening the time-frame required toobtain usable results. The genetics of cereals and grasses havenot diverged greatly throughout evolution, so the geneticinformation derived from the mapping exercise can beexploited across many other species, including rice, wheat,barley, maize, sorghum, sugarcane, Brachiaria and Lolium(rye-grass). Under the umbrella of the SLP, a broader projectlinking work on a number of these additional species hasbeen launched, with support from the Australian Centrefor International Agricultural Research (ACIAR) and theUK’s Department for International Development (DFID).

Populations of these species, grown at different locationsround the world, have been assessed for characteristics relatedto their digestibility and those plant lines with the mostdivergent characteristics have been crossed to produce the


38

progeny required for the identification of genetic markers.The next step is to correlate the segregation of markers inmapping populations with phenotypic variation for relevantcharacteristics in order to identify quantitative trait loci(QTLs)—the segments of DNA responsible for specifictraits. In the longer term, maps of the genome of each cropwill be developed.

Work is most advanced for pearl millet, the species that hasbeen studied longest. The digestibility of the progeny ofcrosses has been found to range widely, suggesting consid-erable potential for identifying genotypes with superior feedquality. Scientists at the UK’s John Innes Centre (JIC) andthe Institute of Grassland and Environmental Research


• International Livestock Research Institute (ILRI):Animal nutrition and production

• International Crops Research Institute for the Semi-Arid Tropics (ICRISAT):Pearl millet and sorghum

• Centro Internacional de Agricultura Tropical (CIAT):Brachiaria pastures

• International Center for Agricultural Research in the Dry Areas (ICARDA):Barley

• Co-operative Research Center for Molecular Plant Breeding (CRC-MPB), La Trobe University,Australia

Perennial rye-grass

• Institute of Grassland and Environmental Research (IGER), UK:QTL mapping (drought tolerance, downy mildew resistance)

• John Innes Centre (JIC), UK:QTL mapping

• Bangor University, UK:QTL mapping (drought tolerance, downy mildew resistance)

• All India Coordinated Pearl Millet Improvement Project (AICPMIP):Field testing and links with farmers

39

Acting Locally

(IGER) have found 11 QTLs for different traits, one ofwhich has proved statistically significant for digestibility.The indications are that it will be possible to increase stoveryield and quality at the same time, without sacrificing toomuch grain yield. The long-term aim is to combine traitsfor fodder quality and quantity with other important agro-nomic traits, such as drought tolerance and resistance todowny mildew, a serious disease of pearl millet. Collaborativeresearch between ICRISAT, JIC and the University of Bangorhas recently led to identification of the gene responsible forthis disease (see Box).

Researchers based at ICRISAT are also investigating therelative contributions of genetic and environmental(management) factors to the quality and quantity ofsorghum and pearl millet residues. As regards genotype,research on sorghum has confirmed that traditional dual-purpose types, which are tall and reach maturity late, havebetter digestibility characteristics than do modern short-duration types, bred mainly for grain. The level of stemsugar and the quantity of stem produced were both positively

Found: The gene for a serious disease of pearl millet

Dr Katrien Devos, at the UK’s John Innes Centre, writes:

“We now have more than 50 microsatellite markers covering the pearl millet genome. These havealready been transferred to ICRISAT and other research institutes for direct use in plant breeding.Recently, we have used these markers to identify a line with good resistance to downy mildew. Thisline has performed extremely well in the glasshouse, and colleagues at ICRISAT are optimisticthat, following field trials, it will prove suitable for release very shortly to public- and private-sector breeders.

“Working with colleagues at Bangor and ICRISAT, we have identified a single gene that isresponsible for downy mildew resistance. Now we will be able to find out how the gene, and hencethe mechanism of resistance, works. Eventually, this should lead to more efficient and environ-mentally friendly methods of controlling the disease.”


40

correlated with digestibility, as also was the persistence ofgreen leaves (the stay-green trait). Similar results wereobtained for pearl millet, with land races producing bothmore and better quality residues than did modern hybrids.Open-pollinated varieties fell between the two, represent-ing a reasonable compromise between grain production andfodder yield and quality. As regards environmental factors,the results obtained so far suggest that their contributionmay be greater in sorghum than in pearl millet. Plantpopulation density had little effect on residue quantity orquality in pearl millet, but it did affect the productivity ofsorghum residues. The level of soil fertility influenced theprotein content of residues and the amount of residueproduced in both crops, while in sorghum stem sugar per-centage was also influenced.

Research on the other species is progressing well. In barley,researchers at ICARDA have identified markers and QTLsfor several digestibility-related characteristics, including aciddetergent fibre, neutral detergent fibre and lignin contents.In perennial rye-grass, researchers at the Co-operativeResearch Center for Molecular Plant Breeding (CRC-MPB)of La Trobe University, Australia have established linksbetween restriction fragment length polymorphisms (RFLPs)and selected digestibility traits, and have also prepared micro-satellite markers. The CIAT team working on Brachiariahas demonstrated variability in digestibility within themapping population derived from crosses.

The pay-offs to this research should be substantial. An ex-ante assessment carried out in 1997-98 for sorghum andpearl millet indicated that a 1% increase in stoverdigestibility—the lowest level thought likely by plantbreeders and nutritionists—will lead to a 6-8% increase inthe production of milk and meat. In India alone this increasewill be worth at least US$ 42 million at the most cautiouslyestimated level of adoption and up to US$ 208 million ifadoption is more widespread (Figure 5). Returns will be

41

Acting Locally

even higher if farmers in Africa and Latin America benefitas well, and higher still if crop production rises in responseto increased amounts of better quality manure and moreefficient animal traction. Securing the widest possible geo-graphical spread of all these benefits will be a major SLPpriority in the years ahead.

The excellent progress made using genetic markers toimprove feed value does not rule out the possibility of alsomaking progress through conventional selection and breed-ing. Indeed, it is still an open question which approach ismost cost-effective. Both approaches need further methodo-logical refinement, including a sharper definition of the traitsthat indicate feed value. For these reasons, the SLP is alsosupporting some work using a conventional breeding

Figure 5. Predicted annual net benefits from research to improve the feed value of pearl millet andsorghum in India

1998 2000 2002

2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026

-4.00

-2.00

0.00

2.00

4.00

6.00

8.00

10.00

12.00

US$ (x106)

Net benefitsDiscounted net benefitsYear

Source: Kristjanson and Zerbini (1999).


42

approach. The crops covered using this approach are cowpea,maize and sorghum and the work is conducted at CIMMYT,IITA and ICRISAT respectively. Ex-ante assessments of theprobable returns to the maize and cowpea components ofthis research are under way (see below).

Towards a virtual SLP

The SLP now includes or works with ten CGIAR centres,six ecoregional consortia and numerous research institutionsacross Africa, Asia and Latin America, in addition to otherCGIAR system-wide programmes. It also sustains partner-ships with an ever broadening circle of specializedinstitutions and other stakeholders in the developed world.The dispersed nature of the SLP and its work requiresdedicated mechanisms to minimize the transaction costs ofresearch, ensure efficient information exchange and promotecollaboration among these partners. Ways need to be foundof ensuring that the SLP’s research as a whole is more effectivethan the sum of its parts.

43

Acting Locally

For this reason, the SLP has decided to design and imple-ment a virtual SLP. Supported by modern, Internet-basedcommunications technologies, the virtual SLP or vSLP willreproduce all the functions of a conventional research station(Figure 6). These functions will be grouped in two majorcomponents: a virtual network and a set of virtuallaboratories. The virtual network, which will be used by allSLP members and partners, will provide a “one-stop shop”for a range of research support functions, including libraryand communications, a coffee or chat room, projectplanning and administration. It will also facilitate the sharingand use of methodologies, data bases and other forms ofinformation, such as project proposals and reports or fundingand budget information. The virtual laboratories, in contrast,will provide settings in which more detailed collaborativeresearch can be undertaken by a group of specialists on agiven topic. Access to these laboratories may if necessary berestricted to partners in the project concerned. Whereasvirtual laboratories will generate results, the virtual networkwill make them available to others.

Figure 6. The virtual SLP

Administration

Map room

Library resources

Croplaboratory

RESTRICTED AREAAUTHORIZED

EMPLOYEES ONLY

NOTICE

Otherlaboratory

room (social)C ffee

Meeting room

Source: SLP (2000a).


44

Virtual laboratories represent a new research paradigm,opening up opportunities for research that was either whollyimpossible or else not feasible on a realistic time-scale amongpartners collaborating in the conventional way. They willchange the nature of collaboration, greatly facilitating thecollection, analysis and interpretation of common data setsfrom multiple locations. And they should allow researchersto generate results with wide applicability.

To implement the vSLP, the SLP has formed a partnershipwith the ILRI programme, Strengthening Partnerships forLivestock Research (SPLR). In early 2000 the SLP and theSPLR programme held a workshop in Addis Ababa, Ethiopiato discuss the concept of the vSLP and plan the way forward.Participants included representatives from the SLP and itsresearch partners in addition to information specialists fromtwo organizations with valuable experience in producingvirtual communication systems for use in developing coun-tries: the Food and Agriculture Organization of the UnitedNations (FAO) and the International Institute for Theoreticaland Applied Physics (IITAP). The United Nations Educa-tional, Scientific and Cultural Organization (UNESCO)also contributed technical expertise and provided partialfunding for the workshop. Participants agreed on the contentand mechanisms of the vSLP and a phased approach for itsimplementation through to the year 2002.

The immediate priorities, which are currently beingaddressed, are to develop an SLP website and to constitutethe network’s basic framework for data management andarchiving. Two data bases are being established in 2000 anda further three will follow in 2001. Directories of SLP partners,general information on projects and programmes, researchoutputs, impact assessments, an inventory of methodologies,a newsletter and a bulletin board are being developed.

To gain experience in virtual laboratory research, theworkshop proposed a pilot project to establish a virtuallaboratory during the year 2000. The laboratory will be

45

Acting Locally

devoted to modeling crop-livestock interactions. Dispersedresearchers will have access to models enabling them to assesscrop-livestock systems in different ecoregions in terms oftheir energy and nutrient flows and other parameters. Thepartners will include the Quito and Lima research locationsof the Centro Internacional de la Papa (CIP), ILRI-Nairobiand ILRI-Niamey, the University of Edinburgh in the UK,Biotechnología Agropecuaria S.A. (BTA) in Chile and theUniversity of Georgia, in the USA.

While virtual networks and laboratories provide greaterefficiency and reduce the cost of information exchangeamong members of a dispersed group, some new costs arenevertheless associated with them. These include trainingin information technology, the provision of standardizedequipment and the hiring of dedicated IT personnel. Tomeet these costs, the SLP has prepared a project proposalwhich it is now submitting to donors for funding.

Ex-ante Impact AssessmentsEx-ante assessments of the impact of research are animportant mechanism through which the SLP establishesits priorities. The primary purpose of these assessments is

Project partners and their strengths

• System-wide Livestock Programme (SLP)Impact-oriented livestock research

• Strengthening Partnerships for Livestock Research (SPLR) (programme of ILRI)Web-based information management and dissemination on livestock-related subjects

• Food and Agriculture Organization of the United Nations (FAO)Virtual communications and information systems for developing countries

• International Institute for Theoretical and Applied Physics (IITAP)Virtual communications systems and laboratories for developing countries


46

to quantify the economic benefits that can be expected froma given project. This ensures that scarce resources for researchcan be allocated wisely, helping to build the confidence ofpotential investors. But the assessments also serve otherpurposes, including assessing the needs of farmers,identifying possible constraints to the adoption of newtechnology and planning the conduct of research amongpotential partners. Involving partners in ex-ante assessmentsis a way of ensuring their ownership of the resulting projectonce it is funded.

Two major ex-ante assessments have been undertaken sincethe SLP’s last biennial report. The first covers ongoingresearch in West Africa to improve the feed value of cowpea,while the second relates to a proposal for research on theuse of maize as food, feed and fertiliser in Eastern andSouthern Africa.

Cowpea in West Africa

Cowpea is widely grown for both grain and fodder in semi-arid West Africa and is also gaining popularity in other partsof the continent. Together with national colleagues, scientistsat IITA and ILRI have developed improved dual-purposevarieties, which are now being disseminated through themulti-partner project described above. In 1999, the SLPcommissioned an ex-ante assessment of the potential impactfrom this work.

As a first step, two workshops were held with farmers todetermine what kinds of impact were important to them.The farmers came from two contrasting areas of northernNigeria: Bichi, a village close to the major market of Kano,where farmers are already familiar with the improved varie-ties, and Minjibir, a more remote village in which mostfarmers still have no experience of improved varieties. Whilemen attended the workshops proper, held at IITA’s Kanostation, their wives were interviewed at home in the villages.

47

Acting Locally

Participants identified no less than eight kinds of benefitfrom improved cowpea. The crop:

• Was a source of income from the sale of beans and/orfodder

• Was a high-quality food for the family

• Provided employment, especially for women and chil-dren, who were hired for such tasks as weeding, harvestingand threshing

• Was a preferred fodder for livestock, at least for farmersnot growing groundnut

• Reduced erosion from wind and rain

• Enriched the soil

• Was a catalyst for joint action, helping farmers raise cashfor shared items such as knapsack sprayers or for regis-tering a cooperative

• Served as an incentive to improve farm managementpractices.

Exercises to rank these benefits in order of importancerevealed differences between men and women and betweenfarmer groups with different market access. Men particularlyvalued the income and food benefits, while women empha-sized home cooking and consumption and the feeding ofsmall ruminants. Farmers from Minjibir stressed the crop’scontribution to their subsistence, while those at Bichi valuedits income-earning potential.

The adoption of improved cowpea varieties is already underway in parts of northern Nigeria. The second step in the ex-ante assessment was therefore to find out more about whois adopting, why they are doing so and how rapidly adoptionis taking place. This information is needed as a baseline fromwhich to predict and measure future adoption and impact.To obtain it, ILRI and IITA conducted a village-level survey


48

of 80 communities in Kano and neighbouring Jigawa States,followed by a more detailed household-level survey inselected areas. The data collected in the course of thesesurveys are still being analysed, but the results are expectedto provide an adequate basis for predicting impact in detail.

The project partners have developed a computer model,CROPGROW-Cowpea, that can predict the grain and for-age yield of cowpea under different agro-ecological andmanagement conditions. The model is currently beingvalidated using data sets from the region and elsewhere. Itspredictions will be linked with data from the adoption sur-veys and with the use of a GIS. The end result will be acomprehensive and detailed picture of the economic, socialand environmental impact that can be expected from thenew technologies.

Maize as food, fodder and fertiliser

Smallholders in Eastern and Southern Africa are respondingto increasing population density and improved market accessby intensifying their mixed crop-livestock productionsystems. Intensification often involves launching a dairyenterprise, which increases the demand on the system forhigh-quality feed. At the same time, the area available forgrazing livestock may be shrinking as cropping expands.These pressures make it imperative to enhance the qualityand quantity of fodder produced on arable land.

Maize is the staple food of 24 million households in theregion. Understandably, research on the crop has tended toemphasize grain production. As a low-quality fodderavailable only seasonally, maize residues have received scantattention from plant breeders and animal nutritionists. Thelittle research conducted on maize stover has focused mainlyon how to better exploit its use to maintain soil fertility. Byemphasizing such factors as the control of weeds and theuse of optimum plant densities, agronomists have also

49

Acting Locally

contributed to the neglect of the maize crop as a source offodder.

The way farmers manage their maize-based mixed crop-livestock production systems is complex. Some practicesenhance the overall productivity and feed value of the maizecomponent, while others detract from it. Many farmers plantmaize at high population densities, then gradually thin thestand down, feeding the thinnings to their animals.Immature maize plants are a high-quality fodder, containingsix times the crude protein (CP) content of dry stover. Weedsfrom maize fields may also be fed to animals—a benefitthat disappears when crops are sprayed with herbicide. Onthe other hand, farmers often apply manure produced partlyfrom feeding maize stover to other, more valuable crops,inducing a transfer of nutrients away from maize fields. Theymay burn residues instead of feeding them or incorporatingthem into the soil. And they may decline to adopt a numberof practices known to improve the feed value of stover, suchas soaking and chopping it before feeding.

The SLP supports a project to develop and disseminatemanagement practices for maize that will increase the pro-ductivity of crop-livestock systems through higher and more


50

stable yields of both grain and fodder while protecting thenatural resource base. Led by the International Wheat andMaize Improvement Center (CIMMYT), the projectcurrently involves national scientists in Kenya, Malawi,South Africa, Tanzania and Zimbabwe.

The ex-ante assessment for this project is well under way.Given the complexity of maize-based systems, a major aimof the assessment is to identify the most effective points atwhich to intervene and the best pathways for overall systemimprovement, bearing in mind all three of the crop’s majoruses. The scientists responsible for the assessment havecharacterized the major maize-livestock production systemsin the five participating countries, conducting detailedstudies on selected systems (see Box). They have estimatedthe importance of each system for a range of parametersthat will determine the impact of future interventions,including the area under maize in each production system,the contribution of maize residues to animal nutrition andsoil fertility and of maize grain to human nutrition, and soon. The scientists are currently evaluating the impact ofpossible interventions to improve fodder quantity and qual-ity. Both the genetic and the management pathways toincreased productivity are under consideration. Themanagement practices covered include planting at highdensities then thinning the crop, stripping leaves for feeding,and chopping and soaking the stover. At a system level, thescientists are evaluating experiences in intercropping maizewith legumes or intercropping different maize varieties forfood and feed. Lastly, they are also assessing the relativemerits of manure, crop residues and commercial fertilisersfor restoring soil fertility, both alone and in combination.

The tasks that remain to be completed include quantifyingthe effects of each intervention and modeling their overallimpact in each production system. This will be done usingexisting crop and livestock simulation models, which willbe linked to ILRI spreadsheet models for estimating

51

Acting Locally

Tanzania: Can the decline in yields be reversed?

FAO statistics show a declining trend for maize yields in Tanzania over the past 20 years. Tanzaniaforms a marked contrast with neighbouring Kenya, where yields have risen by 40% over the sameperiod.

The characterization studies conducted by the project covered two maize-based crop-livestocksystems in the country’s Hai and Lushoto Districts. They led to the selection of Kwalei, a watershedin the environmentally sensitive Usambara Mountains of Lushoto District, as a site for moredetailed research. Kwalei has a relatively low population of improved crossbred dairy cows andpoor market access at present, but the area has considerable potential for development, particu-larly as the market for ecotourism expands. Conducting research there will allow opportunities tocompliment existing activities of the African Highlands Initiative (AHI), which already usesKwalei as one of its benchmark sites.

The studies’ findings confirmed the national picture of declining maize yields and showed why thedecline is happening. Farmers in Kwalei reported that their maize yields had fallen continuouslysince the 1950s, citing declining soil fertility and unreliable rainfall associated with deforestationas the main reasons. In both Hai and Lushoto, land cultivated to maize is being steadily drainedof its nutrients as farmers feed their maize residues to livestock but apply the manure produced toother, more valuable crops, such as coffee and bananas. Few farmers apply commercial fertilisersat present. Severe phosphorus deficiency was evident in farmers’ maize fields.

Possible points of intervention for research at Kwalei include manure handling and compostingtechniques, which are little used at present. The use of fodder trees as a supplement to maizeresidues is a further option well worth exploring. Here as elsewhere, the traditional system clearlyneeds all the help it can get in responding to the challenge of increasing productivity withoutfurther mining the resource base.

economic surplus. The study’s findings will be published inthe ILRI impact assessment series, as well as in an interna-tionally refereed journal.


52

Meeting Our Priorities——Transregional and near-term impactthrough technology transfer ——

As part of its strategy for maximizing the returns toinvestments in global livestock research, the SLP emphasizessupport to projects with a potential for near-term and/ortransregional impact.

As might be expected, the potential for near-term impact isgreatest in the high-potential zones in which research is mostadvanced. In the highlands of Eastern and Southern Africa,the SLP has recognized that potential by supporting ICRAF’sappointment of a technology transfer specialist to build onand extend the results already achieved at Embu, in theKenyan highlands. Most of the work done so far focuses onthe potential for the local transfer of results within Kenya.In this country alone, the planting of Calliandra is thoughtcapable of providing benefits worth US$ 137 millionannually. Other countries in the region in which thistechnology could prove widely applicable include Uganda,Tanzania, Rwanda and Zimbabwe. A start has been madein extending results to these countries, but much more needsto be done.

In the more difficult, drier areas, there is, as yet, lesstechnology available to be transferred. However, a numberof factors hold out hope that impact may be just around thecorner. First, the research paradigm has changed. Theinvolvement of users in a more participatory approach greatlyincreases the chances of success. Second, the isolated effortsof the past have given way to a networking approach inwhich successes and failures can be quickly shared. And third,in regions such as the Sahel, after years of stagnation aneconomic recovery is at last paving the way for the increasedadoption of technology by farmers eager to create a surplusfor market.

53

Thinking Globally

As well as extending technology geographically, much canbe achieved by finding new groups of users in the areas wherea technology is already present. Thus, for example, researchat Embu has shown that goat producers can benefit byfeeding Calliandra just as cattle producers can. Targetingthis additional user group, which is generally less wealthythan cattle owners, will make a useful contribution to povertyeradication.

The technologies being tested through adaptive on-farmresearch will for the most part transfer locally andsubregionally. Tropileche is a good example of the potentialof ecoregional consortia to secure this kind of technologytransfer, as its work with Cratylia in Costa Rica demonstrates.In Eastern and Southern Africa, the work of the AfricanHighlands Initiative (AHI) has been responsible fortransferring a number of technologies besides Calliandra,including the use of climbing beans. Such technologies canmove rapidly between countries with similar ecologies.However, it is more difficult to conceive of the transfer of


54

finished technologies across continents or ecoregions. Eachof the consortia supported by the SLP, for example, facesvery different agro-ecological conditions, making suchtransfers unlikely.

The same is not true of the more upstream researchmethodologies and technologies. The best-bet approach toon-farm research, currently being tested and refined withSLP support in several West African countries, may wellprove more widely applicable in other parts of Africa and inLatin America and Asia. Genetic research to improve thefeed value of pearl millet and sorghum residues could havevery widespread impact indeed. While new varieties of thesecrops developed for India are unlikely to prove immediatelyuseful in the Sahel, the genetic information, expertise andmethods developed through this work will constitute avaluable resource that can be used by all plant breedersworking on the crops worldwide. In addition, the similarityof the pearl millet genome to that of other graminae meansthat the knowledge gained through this work could speedup research on these other crops, bringing forward the dateon which useful results will become available and incurringconsiderable savings in the costs of research.

55

Thinking Globally

The SLP is contributing to work on a number of crops thatare very widely grown, increasing the potential for impactwhen technology is adopted by farmers. Maize is consumedby an estimated 24 million households in Eastern andSouthern Africa alone. The SLP considers support to theCIMMYT-led regional project on this crop to be wellworthwhile because of its high potential for spillover to otherregions of Africa, Asia and Latin America. Similarly, dual-purpose sorghum and pearl millet are important food andfeed crops throughout semi-arid sub-Saharan Africa, wherethey occupy 12.6 and 14.9 million ha respectively—areaseven larger than those in India. Pearl millet is also gainingin importance in Latin America, particularly in Brazil.

The creation of the virtual SLP lays the foundations forconsiderable gains in the efficiency with which technologyand information can be shared and transferred to new users.Indeed, the vSLP will radically alter the way the SLPconducts its business. The laboratory component will makeintensive collaboration possible among widely dispersedresearchers, opening up new lines of global researchpreviously ruled out because of their complexity and highcost. The networking component will make the results ofsuch research widely available, in addition to allowingbroader and more rapid access to research resources such asmodels, software, maps, publications and data sets.

The SLP also supports conventional methods ofdisseminating information, such as conferences, trainingevents and publishing. These methods will remain importantfor transferring research methodologies and results, especiallyfor people unable to connect to the Internet. Often, theSLP provides seed money for a workshop to plan a newproject or define the way forward for an existing one. In1996, the SLP organized an international workshop tofurther the cause of global research on crop residues.

The SLP supports a growing number of projects with a focuson crop-livestock interactions, each of which uses a


56

somewhat different approach to address different aspects ofthis complex topic. Of the three ecoregional projects, forexample, one operates in a single location from which it hasoutreach activities, the second is formally constituted as aconsortium operating directly under the SLP, while the thirdis a set of discrete national projects in a larger set of countrieslinked through networking. A recently launched SLP projectwill examine the potential of these projects for transregionalimpact, in terms of both the technologies tested with farmersand the operational modes used by researchers. The projectshould lead to a better understanding of global crop-livestocksystems and the factors driving their evolution, which inturn will facilitate the setting of research priorities and thetargeting of solutions to wider recommendation domains.

In conclusion, the SLP’s quest for near-term andtransregional impact through technology transfer is alreadypaying dividends. We are confident that the years aheadwill provide increasing evidence of impact.

57

Thinking Globally

The Challenges AheadDespite the achievements of the past 2 years, the most seriouschallenge facing the SLP is that of fund raising. Fundingwithin the CGIAR as a whole not only remains extremelytight but is also increasingly difficult to predict. AtInternational Centres Week (ICW) in 1999, the CGIARFinance Committee decided to set aside US$ 5 millionannually for allocation to system-wide programmes. On thebasis of its record in developing effective and relevant projectsthat enjoy broad-based support from participants and thatcan demonstrate near-term impact, the SLP will be a strongcontender for a share of these funds.

The SLP also faces other challenges. While the number ofprojects it supports has increased rapidly, it remains essentialto ensure that the whole research effort is greater than thesum of its parts. Linking related research activities andderiving transregional benefits from research were keyobjectives at the SLP’s foundation. In the future, greateremphasis on transregional analysis will be needed as a meansof maximizing impact.

The tight funding environment and the need to achievetransregional impact make efficiency and synergy in theSLP’s operations and research crucial to its success. The SLPnow involves 10 CGIAR centres and a host of other partnersin ecoregional, system-wide and national programmes. Thetransactional costs of working with these partners byconventional means are high. One solution is to develop a“virtual SLP”—an initiative already being pursued (seeabove). Further opportunities for innovative ways of cuttingcosts and increasing impact will also need to be sought.

As the SLP’s projects and project components deliver results,it will become increasingly necessary to assess impact. MuchSLP research forms part of existing CGIAR centreprogrammes and as such is already subject to centre-specific


58

review mechanisms. However, there will be a strong casefor evaluating the SLP’s effectiveness as a whole at somestage, preferably once the transregional benefits start tomaterialize.

59

Annexes

Sources

CIAT (Centro Internacional de Agricultura Tropical) 2000. The Tropileche Experience,Phase 1: Achievements and Lessons in Addressing the Needs of Smallholder CattleProduction in Latin America. Cali, Colombia.

Gintzburger, G., Bounejmate, M., Agola, C. and Mossi, K. (eds) 2000. Production andUtilization of Multi-purpose Fodder Shrubs and Trees in West Asia, North Africaand the Sahel. International Center for Agricultural Research in the Dry Areas(ICARDA), Aleppo, Syria, and System-wide Livestock Programme (SLP), Addis Ababa,Ethiopia.

Grandi, J.C. (ed.) 1996. L’évolution des systèmes de production agropastorale par rapportau développement rural durable dans les pays d’Afrique Soudano-Sahélienne. CollectionFAO: Gestion des Exploitations Agricoles, No.11. Food and Agriculture Organizationof the United Nations (FAO), Rome, Italy.

Holmann, F., Lascano, C. and Kerridge, P. 1998. Progress Report 1998: TropilecheConsortium. Report presented at ILRI’s Annual Programme Meeting, 14-18 September1998. ILRI, Addis Ababa, Ethiopia.

ICRAF (International Centre for Research in Agroforestry), KARI (Kenya AgriculturalResearch Institute) and ILRI (International Livestock Research Institute) 2000.Utilization of Forage Legume Biodiversity for Dairy Production and Natural ResourcesManagement in the East and Central African Highlands. Progress Report to the System-wide Livestock Programme. ILRI, Addis Ababa, Ethiopia.

INRA (Institut National de la Recherche Agronomique) 1999. Shrub Project, Productionand Utilization of Multi-purpose Fodder Shrubs and Trees in West Asia, North Africaand the Sahel: Progress Report. Settat, Morocco.

Kristjanson, P., Tarawali, S. and Okike, I. 1999. Farmer Impact Workshops for EvaluatingImpact of Improved Food and Fodder Cowpea Varieties in Northern Nigeria: A Reportto the System-wide Livestock Programme. ILRI, Nairobi, Kenya.

Kristjanson, P. and Zerbini, E. 1999. Genetic Enhancement of Sorghum and Millet ResiduesFed to Ruminants: An Ex-ante Assessment of Returns to Research. ILRI ImpactAssessment Series No. 3. ILRI, Nairobi, Kenya.


60

SLP (System-wide Livestock Programme) 1999. Genetically Improving Grain and FodderYields of Cowpeas: Measuring the Potential Impact of the Research. 1999 Annual Reportfor SLP: Overview of Activities. ILRI, Addis Ababa, Ethiopia.

SLP (System-wide Livestock Programme) 1999. The Maize Crop as Food, Fodder andFertiliser in Intensifying Crop-livestock Systems in East and Southern Africa. 1999Report to the SLP: Project Summary. ILRI, Addis Ababa, Ethiopia.

SLP (System-wide Livestock Programme) 2000. Intensification of Integrated Crop-livestockSystems in the Dry Savannas of West and Central Africa: SLP Submission to theInternational Fund for Agricultural Development (IFAD). ILRI, Addis Ababa, Ethiopia.

SLP (System-wide Livestock Programme) 2000. Ex-ante Impact Assessment of Maize asFood, Fodder and Fertiliser in Intensifying Crop-livestock Systems in East and SouthernAfrica: Status, March 200. ILRI, Addis Ababa, Ethiopia.

SLP (System-wide Livestock Programme) 2000. Progress Report. Paper presented at theLivestock Programme Group Meeting, 29-31 May 2000, Addis Ababa, Ethiopia.

SLP (System-wide Livestock Programme) 2000a. Planning the Development of a VirtualSLP. Workshop report of the UNESCO/SLP Workshop, 28 February-1 March 2000,ILRI, Addis Ababa, Ethiopia.

Tarawali, S.A., Singh, B.B., Gupta, S.C., Tabo, R., Harris, F., Nokoe, S., Fernandez-Rivera,S., Bationo, A., Manyong, V.M., Makinde, K. and Odion, E.C. 2000. Cowpea as a KeyFactor for a New Approach to Integrated Crop-livestock Systems Research in the DrySavannas of West Africa. Paper presented at the Third World Cowpea ResearchConference, 4-7 September 2000, IITA, Ibadan, Nigeria.

Tarawali, S.A., Smith, J.W., Hiernaux, P., Singh B.B., Gupta, S.C., Tabo, R., Harris, F.,Nokoe, S., Fernandez-Rivera, S. and Bationo, A. 2000. Integrated Natural ResourceManagement: Putting Livestock in the Picture. Paper presented at the Integrated NaturalResource Management Meeting, 20-25 August 2000, Penang, Malaysia.

Thorne, P.J. 1999. Maize Food and Forage Production in Smallholder Mixed FarmingSystems: A Case Study from Tanzania. Consultant’s report. SLP-ILRI, Addis Ababa,Ethiopia.

University of Durham, IFDC (International Fertiliser Development Centre), ICRISAT(International Crops Research Institute for the Semi-Arid Tropics), ILRI (InternationalLivestock Research Institute) and IITA (International Institute of Tropical Agriculture)

61

Annexes

2000. Improving Crop-livestock Systems in the Dry Savannahs of West and CentralAfrica: Progress Report, March 2000. ILRI, Addis Ababa, Ethiopia.

Zerbini, E., Hash, C.T. and Reddy, B.V.S. 2000. Genetic Enhancement of Feed Qualityand Quantity in Sorghum and Pearl Millet: Semi-annual Report, October 1 to March31. ILRI-ICRISAT, Hyderabad, India.

Zerbini, E., Hash, C.T., Fernandez, S., Baum, M., Ceccarelli, S., Thome, J., Lascano, C.,Spangenberg, G., Forster, J., Stone, B. and Lam, B.T. 1999. Development and Use ofMolecular Genetic Markers for Enhancing the Feeding Value of Cereal Crop Residuesfor Ruminants. Australian Restricted Project Grant to International Research Centers,Progress Report, 1 July 1998-31 May 1999. ILRI, Nairobi, Kenya.


62

SLP-supported Projects and their Status

1. The original three ecoregional projects established or supported during Phase I are:

• Utilization of forage biodiversity for dairy production and natural resource managementin the African highlands

Lead centre: ICRAFMain partners: ILRI, AHI and national institutions in Kenya.

• Improving legume-based feeding for smallholder dual-purpose cattle production intropical Latin America and the Caribbean

Lead centre: CIATMain partners: ILRI and national institutions in Costa Rica and Peru.

• Production and utilization of multi-purpose shrubs in West/Central Asia-North Africaand the Sahel

Lead centre: ICARDAMain partners: ILRI and national institutions in Jordan, Morocco, Pakistan, Syria,Tunisia, Burkina Faso, Mali, Niger and Senegal.

2. In 1998-99 a further group of projects was funded wholly or in part under Phase II.These are:

• Improving crop-livestock systems in the dry savannas of West AfricaLead centre: IITAMain partners: ICRISAT, IFDC, ILRI, University of Durham (CORD) and nationalinstitutions in Niger, Nigeria and Mali.

• Development and use of molecular genetic markers for enhancing feed value of cropresidues

Lead centre: ILRIMain partners: CIAT, CIMMYT, ICARDA, ICRISAT, Bangor University, IGER,

JIC, CRC-MPB, AICMIP and various other national institutions.

• The maize crop as food, feed and fertiliser in intensifying crop-livestock systems inEastern and Southern Africa

Lead centre: CIMMYT

63

Annexes

Main partners: ILRI and national institutions in Kenya, Malawi, South Africa,Tanzania and Zimbabwe.

• A set of ex-ante impact assessments of research on crop residues in mixed farming systemsLead centre: ILRIMain partners: IITA, CIMMYT and national institutions in Kenya, Nigeria, Tanzaniaand Zimbabwe.

3. In 1999 the following additional projects were approved for support:

• Improving productivity in mixed crop-livestock farming systems in South AsiaLead centre: ICRISATMain partners: ILRI and national institutions in Bangladesh, Bhutan, India, Nepal,Pakistan and Sri Lanka.

• Enhancing livestock productivity while protecting mountain ecosystemsLead centre: CIPMain partners: ICIMOD, ILRI and national institutions in China, Ethiopia, Nepaland Peru.

• Differential impacts on small-scale and large-scale livestock (including dairy) producersof expanded use of feed concentrates: case studies in Bangladesh, Kenya and thePhilippines

Lead centre: IFPRIMain partners: ILRI and national institutions in Bangladesh, Kenya and thePhilippines.

• Research and development of smallholder livestock production in Central AsiaLead centres: ICARDA, IFPRI and ILRIMain partners: selected national institutions in Central Asia and the Caucasus.

• Human population growth and poverty mapping: implications for natural resourcemanagement

Lead centre: ILRIMain partners: CIAT, ICRAF and the Rockefeller Foundation.

• The role of agroforestry and livestock strategies in building assets and improvinglivelihoods

Lead centre: ICRAFMain partners: ILRI, AHI and national institutions in Kenya and Tanzania.


64

4. Lastly, a further group of projects are under development. These are:

• Participatory development of legume-based technologies for intensifying livestock systemsin Latin America

Lead centre: CIATMain partners: ILRI and national institutions in Central America and the Caribbean.

• Wheat and weeds, food and feed in the highlands and mountains of sub-Saharan Africaand Latin America and the Caribbean

Lead centre: CIMMYTMain partners: ILRI, CIP and national institutions in Ethiopia and Bolivia.

• Food/feed systems and improved livelihoods of the poor in rainfed lowland and uplandareas of Southeast Asia

Lead centre: ILRIMain partners: IRRI, CIAT and national institutions in the Philippines, Indonesia,Vietnam and Cambodia.

• The SLP as a virtual network (vNetwork) and virtual laboratories (vLabs)Lead centre: ILRIMain partners: FAO, IITAP and other SLP collaborators.

65

Annexes

Financial Statement

Income and Expenditure, 1999/2000

This is the list of investors who support the SLP directly. We are grateful not only to thembut also to the investors who support the SLP indirectly through unrestricted funding tothe CGIAR centres.

Income 1999 2000

Brought Forward — 470 212World Bank1 1 200 000 —BMZ 228 000 202 000Japan 100 000 100 000Switzerland 291 000 263 000Netherlands — 117 000ACIAR 98 0002 —

Total 1 917 000 1 152 212

Expenditure

Programme Management3 184 192 250 000Programme Development4 85 168 161 269Project Support5 1 177 428 740 943

Total 1 446 788 1 152 212Carried forward from 1999 470 212 —

Notes:

1 World Bank funds granted by the CGIAR’s Finance Committee at the Mid-term Meeting of May 1999.Expended over 1999 and 2000.

2 The 1999 portion of a 3-year project budget that flows directly to project partners.3 Programme management expenditures include operating the SLP Secretariat, meetings of the LPG,

publications, project evaluation missions, etc.4 Project development expenditures include project design workshops and ex-ante impact assessments to

quantify pay-off and prioritize research opportunities.5 Grants made to centres on their own and partners’ behalf for executing projects approved by the LPG.


66

Actual and Projected Project Support (Research) in 1999/20001

(Million US$)

Portfolio of Projects 1999 2000

Enhancing livestock productivity while protecting mountainecosystems (2000-2001) 0.180

Research and development of smallholder livestock systems inCentral Asia (2000-2001) 0.300

Differential impacts on small-scale and large-scale livestock anddairy producers of expanded use of feed concentrates: case studiesin Bangladesh, Kenya and the Philippines (2000-2001) 0.320

Human population growth and poverty mapping: implicationsfor natural resource management (1999-2000) 0.086

Utilization of forage biodiversity for dairy production and naturalresource management in the African highlands (1997-2000) 0.050

Improving legume-based feeding for smallholder dual-purpose cattleproduction in tropical Latin America and the Caribbean (1997-2000) 0.080

Production and utilization of multi-purpose shrubs in West/CentralAsia-North Africa and the Sahel (1997-2000) 0.071

Improving crop-livestock systems in the dry savannas of West Africa(1999-2001) 0.292 0.184

Improving productivity in mixed crop-livestock farming systemsin South Asia (1999-2001) 0.200

Development and use of molecular genetic markers for enhancingfeed value of crop residues (1999-2001) 0.181 0.150

A set of ex-ante impact assessments of research on crop residuesin mixed farming systems (1999-2000) 0.117 0.064

The maize crop as food, feed and fertiliser in intensifying crop-livestock systems in Eastern and Southern Africa (1999-2001) 0.100

Total: 1.177 1.1982

Notes:

1 1999, actual support; 2000, projected support.2 US$ 0.74 million funded out of 2000 contributions and US$ 0.458 million funded from Brought Forward.

67

Annexes

Centre Addresses and Contacts

CIAT Dr. Carlos LascanoCentro Internacional de Agricultura TropicalApartado Aéreo 6713Cali, Colombia

Tel: (57-2) 445-0000Fax: (57-2) 445-0073E-mail: [email protected]

CIMMYT Dr. Dennis FriesenCentro Internacional de Mejoramiento de Maiz y TrigoCIMMYT-KenyaPO Box 25171Nairobi, Kenya

Tel: (254-2) 632-054/632-206/630-165Fax: (254-2) 631-499/630-164E-mail: [email protected]

CIP Dr. Roberto QuirozCentro Internacional de la PapaApartado 1558Lima 12, Peru


ICARDA Dr. Luis IniguezInternational Center for Agricultural Research in the Dry AreasPO Box 5466Aleppo, Syria

Tel: (963-21) 2-225-517Fax: (963-21) 2-225-105E-mail: [email protected]


68

ICRAF Dr. Steven FranzelInternational Centre for Research in AgroforestryUnited Nations AvenuePO Box 30677Nairobi, Kenya


ICRISAT Dr. Barry ShapiroInternational Crops Research Institute for the Semi-Arid TropicsICRISAT-BamakoBP 320Bamako, Mali

Tel: (223) 223-375/227-707Fax: (223) 228-683E-mail: [email protected]

IFPRI Dr. Chris DelgadoInternational Food Policy Research Institute2033 K Street, N.W.Washington DC 20006, USA


IITA Dr. Dyno KeatingeInternational Institute of Tropical AgriculturePMB 5320Ibadan, Nigeria


69

Annexes

ILRI Dr. Jimmy Smith, Programme CoordinatorSystem-wide Livestock ProgrammePO Box 5689Addis Ababa, Ethiopia

Tel: (251-1) 615-867Fax: (251-1) 611-892/614-645E-mail: [email protected]

IRRI Dr. Mercedita SombillaInternational Rice Research InstitutePO Box 933Manila, The Philippines


Credits

Compilation, editing:Simon Chater, Green Ink Publishing Services Ltd, UKwith Clare Robinson, John Innes Centre, UK

Design and layout:Christel Blank, Green Ink Publishing Services Ltd, UK

Photos:James Bedding/Panos Pictures: p. 1; Piers Benatar/Panos Pictures: p. 36 (above); Moustapha Bounejmate/ICARDA: p. 19 (above and below), p. 20, p. 22; Arabella Cecil/Panos Pictures: p. 11; Dave Elsworth/ILRI: p. 4, p. 8, p. 18, p. 28, p. 32, p. 35 (above and below), p. 36 (below), p. 40, p. 53 (below), p. 54, p.56(right), p. 58; Jeremy Hartley/Panos Pictures: p. 47; Fred Hoogervorst/Panos Pictures: p. 23; JeremyHorner/Panos Pictures: p. 56 (centre); Daniel O’Leary/Panos Pictures: back cover; Luis Fernando Pino/CIAT: p. 13 (below); Giacomo Pirozzi/Panos Pictures: p. 25, p. 33; Betty Press/Panos Pictures: p. 10, p.48, p. 49, p. 50; Clare Robinson/JIC: p. 39; Marcus Rose/Panos Pictures: p. 42 (left); Nathan Russell/CIAT: p. 13 (above), p. 15, p. 16; Chris Sattlberger/Panos Pictures: p. 42 (right); Sean Sprague/PanosPictures: p. 53 (above) and front cover; Ray Wood/Panos Pictures: p. 3; E. Zerbini/ILRI: p. 56 (left)

Printing and binding:Pragati Offset Ltd, India, under the supervision of Sue Hainsworth

About the CGIAR

The Consultative Group on International Agricultural Research (CGIAR) is a consortiumof over 50 public- and private-sector bodies that provide funding for 16 internationalagricultural research centres and several programmes linking their activities, includingthe SLP. Founded in 1971, the CGIAR is sponsored by the Food and AgricultureOrganization of the United Nations (FAO), the United Nations Development Programme(UNDP), the United Nations Environment Programme (UNEP) and the World Bank.

The CGIAR’s mission is to contribute to food security and poverty eradication indeveloping countries through research, partnership, capacity building and policysupport. In pursuit of this mission, the CGIAR focuses on five major research thrusts:increasing productivity, protecting the environment, saving biodiversity, improvingpolicies and strengthening national research. It collaborates with a wide range ofpartners, especially national agricultural research systems, advanced researchinstitutions in the North and the South, universities, the private sector, non-governmentorganizations and farmers’ associations.

Centres supported by the CGIAR

CIAT Centro Internacional de Agricultura TropicalCIFOR Centre for International Forestry ResearchCIMMYT Centro Internacional de Mejoramiento de Maiz y TrigoCIP Centro Internacional de la PapaICARDA International Center for Agricultural Research in the Dry AreasICLARM International Center for Living Aquatic Resources ManagementICRAF International Centre for Research in AgroforestryICRISAT International Crops Research Institute for the Semi-Arid TropicsIFPRI International Food Policy Research InstituteIIMI International Irrigation Management InstituteIITA International Institute of Tropical AgricultureILRI International Livestock Research InstituteIPGRI International Plant Genetic Resources InstituteIRRI International Rice Research InstituteISNAR International Service for National Agricultural ResearchWARDA West Africa Rice Development Association

System-wide Livestock Programme (SLP)International Livestock Research Institute (ILRI),

PO Box 30709, Nairobi, Kenya

CGIAR

SLPSLP

cgiar system-wide livestock programme · compared. it should also greatly speed up the development...

Documents