SATrends Issue 48                                                                                                                November 2004

  • Do relief programs really relieve?
  • Micro-dose d’ engrais pour la prospérité des petits paysans du Sahel
  • Minding your N and P!
  • Testing seed health: Getting the numbers right
  • Formation of a Biopesticides Research Consortium
  • ICRISAT names its Cytoplasmic Male-Sterility system
  • 1. Do relief programs really relieve?

    Zimbabwe’s smallholder farm sector faces many problems – droughts, inflation, unemployment, AIDS... The government and NGOs have launched large-scale relief programs, distributing free seed and fertilizer to assist the rural poor. ICRISAT was asked by several donors to help assess the impact of these programs – and ultimately, suggest ways to make relief programs more efficient.

    The study surveyed over 3200 farm households in 19 districts across Zimbabwe. The results were then discussed at a series of meetings with national and district authorities, extension staff, NGOs, and donors. The study and the subsequent discussions highlighted several key issues.

    First, relief seed was welcome but not essential. Despite two consecutive droughts, most farmers did have seed for planting, either saved from their own harvest or obtained through informal markets or social networks. In contrast to common perceptions, relief seed did not lead to an increase in planted area. Some of the relief seed was of poor quality, failing to meet national standards for germination, physical purity, or genetic purity. Some seed was poorly labeled, or consisted of varieties unknown to farmers. Some farmers received seed of crops they would not normally grow.

    Again contrary to common perceptions, fertilizer distribution can give very high payoffs even in drought-prone areas, provided farmers know how and when to apply it. In 2003/04, about 160,000 farmers received 25 kg each of nitrogen fertilizer, along with specific advice on its application. Average yield gains were 30-50%; and remarkably, almost every recipient achieved significant gains.

    To improve efficiency, two additional issues need to be addressed. One is targeting. In theory, relief assistance is targeted at the poorest or most vulnerable households. But the selection criteria are often impractical to implement; so beneficiaries are not necessarily the poorest in the community. Another issue is duplication of effort. Multiple NGOs often operate in the same area, leading to the delivery of multiple input packages to the same households.

    The study developed a set of guidelines for designing and implementing NGO relief programs. The guidelines discuss the nuts and bolts of relief distribution, as well as the background development issues. They discuss various innovations such as seed fairs and seed/input vouchers, more practical criteria for selecting beneficiaries, planning and implementation, monitoring methods; and even more fundamental questions, such as whether relief investments in other inputs (fertilizer, draft power) can generate better returns than investments in seed distribution.

    For more information contact David Rohrbach, Principal Scientist (Economics), email:

    2. Micro-dose d’ engrais pour la prospérité des petits paysans du Sahel

    La production du mil et du sorgho,principales cultures des populations du Sahel,est faible due aux contraintes biophysiques et socio-economiques. Des tests au Niger ont montré que la productivité de ces céréales s’ améliore par la fertilisation par micro-dose ( application de petites doses d’ engrais au poquet). Avec la micro-dose, la quantité optimale d’ engrais par hectare n’ est  que de 20 kg DAP (18-46-0) ou 60 kg  NPK (15-15-15), correspondant respectivement à 2 g DAP par poquet et 6 g NPK par poquet. D’ autres mesures d’ accompagnement consistent à faciliter l’ accés des producteurs au crédit et aux engrais et à mieux valoriser leur production. Ce système appélé le « warrantage »est développé par le Projet Intrants FAO..

    C’ est à partir de ces expériences que le projet USAID TARGET fertilisation par micro-dose est développé parl’ ICRISAT et ses partenaires, TSBF-CIAT,IFDC, Projet Intrants FAO,les ONGs,INERA, Burkina Faso, IER, Mali, et  INRAN, Niger. Ce projet est financé par l’ USAID. Dans ces trois pays où la technologiea été testée , l’ augmentation des rendements du sorgho et du mil a varié entre 43 à 120 % et les revenus des producteurs ont augmenté de 52 à 134 %. De 2002 à 2003, le projet a atteint près de 15000 menages.

    Poursuivant les efforts de dissemination de cette technologie de micro-dose, des champs écoles d’ un hectare chaque ont été installés sur 4 sites, à Fada Zeno et Konkorido dans la région de Dosso et Guidan Idder et Doguerawa dans la région de Tahoua, Niger. Un groupe de 20 producteurs stagiaires par site a conduit les activités des champs écoles avec l’ appui technique du projet.

    Des Journées Portes Ouvertes (JPO) ont été  organisées du  27 - 30 septembre 2004. Le nombre des participants a varié entre 100 et 550. Des visites des champs et des magasins de « warrantage » ont été faites sous le guide des producteurs stagiaires. Ces JPO ont permis aux villages environnants de se rapprocher des organisations des producteurs des sites où les JPO ont été organisées et de s’ initier à la démarche de création des groupements et des organisations des producteurs en vue de pratiquer la micro-dose, le « warrantage » et les activités génératrices de revenus. L’approche des champs écoles est efficace pour exposer les producteurs aux technologies prometteuses et ainsi accelerer l’ adoption des ces technologies.

    Contacter : Dr Ramadjita Tabo, chercheur principal, ICRISAT ; Niamey, Niger email :
    3. Minding your N and P!

    Nitrogen (N) and phosphorus (P) are two major nutrients required for a high crop yield. When managed inappropriately N becomes a source of pollution and concern. Offshore oxygen depletion in the Atlantic Ocean adjoining Europe and USA has been linked to nitrates (NO3) from agricultural sources (nitrates feed algae which in turn deplete the N in water, making it unsafe for fish). It is also highly likely that N from sewage and related sources is added to the total N pool. The presence of high levels of NO3 (>45 ppm unsafe for humans) in the wells near villages are very likely due to mismanagement of solid (excrements of animal and humans, crop biomass stored/used as fuel and fodder) and liquid (animal urine, sewage water) wastes.

    Phosphorus is a globally scarce element and its supply is dwindling. It is estimated that between 1950 and 2000 about 250 megatones of P reached global water bodies, much of it through agricultural products (Gumbo et al. 2002).

    Both N and P can potentially be recovered and brought back to croplands by addressing the existing anomalies in the N and P cycles. This will require recycling of solid and liquid wastes. Microbiologically safe and low-cost technologies for recycling exist, but will require application research for different situations. Dry composting toilets, modified septic tanks, biogas production using cattle and human excrements and constructed wet lands (root-zone cleaning system) are some of the proven sanitation technologies that can be employed. These would also provide significant quantities of crop nutrients. Research to develop nutrient efficient crop varieties that can meet their N and P needs through organic sources will also be required to reduce fertilizer use.

    In a hypothetical case study in the Adarsha Watershed of village Kothapally in Andhra Pradesh, application of the sanitation technologies was assessed to produce biogas for 100 of the 274 households, provide compost for 10% of the 465 ha village area, generate additional water for the village fish-pond and an additional 1 t of fish (an item of nutritional security). If part of the recovered water is used for composting herbs, it will provide biopesticides to protect 40% of the crops in the village. Input cost of these technologies could be recovered within first two years of implementation. Additionally, the improved sanitation and health would be a bonus. The same model could apply to urban and peri-urban areas, wherever land area is not a constraint.

    For more information contact
    4. Testing seed health: Getting the numbers right

    Assessment of seed health requires laboratory testing of a sample of seeds for the presence of known pathogens. Since any chosen sample of seeds is only a fraction of the parent seed lot, there will always be some degree of uncertainty in sample estimates of inherent pathogen incidence. Choosing too small a sample can lead to highly uncertain, hence practically unusable, estimates. Choosing too large a sample is a waste of costly resources. It is therefore important that the optimal sample size be chosen. Using available pathogen incidence data from five sorghum accessions, each accession tested for incidence of 15 pathogens on a lot of 1050 seeds, parametric and Monte Carlo simulation approaches were used to determine/confirm the optimal sample size. The parametric approaches were based on binomial distribution for pathogen incidence.

    Parametric approaches require an a priori specification of levels of tolerable uncertainty. With uncertainty defined in terms of Type-1-error-probability of 0.05 (Type-1 error refers to rejecting something that is true) for the sample estimate of pathogen incidence to differ from population value by not more than 0.05, a sample of 400 seeds is found to be optimal. This is in line with the recommendations of the International Seed Testing Association (ISTA). Investigations further revealed that a sample of 400 seeds delivers sample estimates with a coefficient of variation (CV) of 15%. Another parametric approach we used was based on statistical power considerations. Power is the probability of not accepting what is false. Setting power at 0.80, and specifying uncertainty as before, requires 1050 seeds to be tested. To achieve 90% power, one needs 1300 seeds. These are expected results – less is the uncertainty we seek in our inferences, more is the effort we need to invest. These parametric approaches do not use actual data but are solely based on the a priori assumption of binomial distribution.

    Extensive Monte Carlo simulations were performed using the actual pathogen incidence data to confirm the above parametric-approach results. Five thousand random samples (without replacement) of sizes n=50, 75, 100, 125, 150, 175, 200, 300, 400, 500, 600 were drawn from the (parent) lot of N=1050 seeds. For each such sample, pathogen incidence was estimated and compared with that obtained from N seeds using a c2 test of goodness of fit. A sample of 200 seeds was found optimal as this provided the closest fit, with median error in pathogen incidence estimation never exceeding 0.022 over the 5000 samples. Use of 200 randomly selected seeds therefore seems adequate. This represents a 50% reduction in cost compared to the ISTA recommended sample of 400 seeds.

    For more information please contact

    5. Formation of a Biopesticides Research Consortium

    Mighty Microbes Manage Pests

    In the last three decades, many countries benefited from the use of chemical pesticides to increase food production, but associatedproblems are now surfacing – insecticide resistance, resurgence of secondary pests, environmental pollution, and their residues in food, drinking water and even in milk. The solution seems to lie with biopesticides, which have the potential to replace chemical pesticides. This potential is being fully harnessed by organic farmers, who are now known globally for not just effectively managing insect pests, but for obtaining yields higher than those of conventional farmers. Although such claims need scientific scrutiny, the fact remains that biopesticides can help us manage insect-pests more efficiently.

    ICRISAT and national partners have been working on managing inset-pests and diseases with microorganisms.Realizing the power of partnerships, a dialogue with the private sector (PS) Biopesticide Industry in India was initiated through two brainstorming meetings (04 May and 08 July 2004), resulting in the endorsement of the ICRISAT-Private Sector Biopesticides Research Consortium (BRC)

    The BRC will enhance the use of biopesticides by farmers. The partnership research will validate protocols for low-cost, commercial-scale production of microbial biopesticides developed at ICRISAT, and will promote agricultural practices that enable protection of crops at a low cost.

    ICRISAT will transfer to the members of the consortium, the technologies for the preparation of biopesticides found to be successful. It will also share the agricultural practices that can prevent pest attack on crops.

    Picture: IPM-Consortium

    Over the last five years, ICRISAT has developed a protocol to prevent the attack of the pod borer, Helicoverpa armigera, on pigeonpea and cotton. Up-scaled technologies for the production of two biopesticides that kill the larvae of Helicoverpa (legume pod borer or cotton bollworm) are being fine-tuned.  While one uses the bacterium Bacillus subtilis (strain BCB-19), the other uses the fungus Metarrhizium anisopliae, both developed at ICRISAT. HNPV (Helicoverpa Nuclear Polyhedrosis Virus), another biopesticide which kills Helicoverpa, will also be available to partners.

    These biopesticides are eco-friendly and are pro-poor. ICRISAT’s trials usingthe two biopesticides in the fields of 17 farmers from the Kothapally village, Andhra Pradesh, India during 2003-04 have shown that total cost of pesticide application can be reduced to US$ 58 per ha per season, compared to over US$ 172 spent on chemical pesticides by cotton farmers.

    Womens’ self-help groups at Kothapally have been trained to produce and supply a wash of Glyricidia and neem leaves which could be sprayed on the crops as additional protection. Another practice to control Helicoverpa is the growing of trap crops

    Caption: Biopesticides kill pests the eco-friendly way

    The response of the private sector to the BRC is encouraging, and to date, fourteen established or start-up biopesticide manufacturers have agreed to become members.

    For further information contact

    6. ICRISAT names its Cytoplasmic Male-Sterility system

    The Barwale CMS System in Pigeonpea

    Pigeonpea [Cajanus cajan (L.) Millsp] is an important food legume of the tropics and semi-arid tropics. It is annually grown on over 4 million ha under subsistence agriculture by resource-poor farmers.The crop is multipurpose, providing not only food, fodder and fuel wood, but also improved soil fertility and structure.Due to changing rural social and environmental factors, the importance of pigeonpea in rainfed dry land agriculture is on the increase.According to one estimate, the pigeonpea area and production are recording an annual growth rate of 2% in recent years.However, inspite of serious breeding efforts, the genetic improvement through pure line breeding has not effectively increased productivity of the crop, which has remained unchanged at around 600-700 kg ha-1.

    Unlike other legumes, pigeonpea is often cross-pollinated, exhibiting 25-30% natural out-crossing.Pigeonpea breeders at ICRISAT developed a genetic male-sterility-based (GMS) technology to exploit hybrid vigor. Using this technology ICRISAT developed the world’s first pigeonpea hybrid, ICPH 8, which recorded 25-30% yield advantage in farmers’ fields. However, this technology could not become popular due to difficulties associated with the genetic nature of male-sterility, which restricts large-scale seed production of female parents and hybrids.To overcome the constraint, ICRISAT and its partners initiated research to develop a cytoplasmic-nuclear male-sterility (CMS) system.  In the past 10 years three CMS systems have been developed using three different wild relatives of pigeonpea.  Among these the CMS derived from a cross involving Cajanus cajanifolius (a wild relative of pigeonpea) and a cultivated line was perfect and stable.  The female-fertility was normal with good pod set from cross pollination.

    Dr Barwale, center, in a CMS field.

    The fertility restoration systems, a pre-requisite for producing fertile hybrids, are also excellent in these male-sterile sources.The pollen production and pod set are similar to those of the existing cultivars.In a short period of two seasons, over 200 fertility-restoring lines with diverse agronomic traits have been selected. Overall, this is a perfect CMS system with all necessary qualities and it has a great potential for developing commercial hybrids in pigeonpea.


     Dr Barwale receiving a memento from Dr Dar.

    Dr BR Barwale, Founder Chairman of Maharastra Hybrid Seed Company (MAHYCO), Jalna, India, is a World Food Prize winner, a great friend of ICRISAT, and the biggest supporter of the CMS system derived from C. cajanifolius. On 15 December, Dr William D Dar, Director General, ICRISAT, proudly named this system the ‘Barwale CMS System in Pigeonpea’.

    For further information contact