FARMER’S PARTICIPATORY PLANT BREEDING

Introduction
Participatory plant breeding (PPB) as an additional, complementary and powerful strategy for advancing the rights and interests of farmers.¹ The International Treaty on Plant Genetic Resources for Food and Agriculture is the first legally binding international agreement that explicitly recognizes farmers’ rights. There are, of course, numerous potential complementary ways to promote farmers’ rights.

Participatory Plant Breeding
Participatory plant breeding (PPB) is the process by which farmers are routinely involved in a plant breeding programme with opportunities to make decisions throughout. Farmers’ involvement in PPB can take many forms: defining breeding goals and priorities; selecting or providing sources of germplasm; hosting trials on their land; selecting lines for further crossing; discussing results with the scientists; planning for the following year’s activities; suggesting methodological changes; and multiplying and commercializing the seed of the selected lines. Participatory variety selection (PVS) refers to processes whereby farmers are involved in selecting lines that they judge to be most appropriate for their own uses from among a range of fixed (stable) lines that are being field tested. PPB generally involves a higher and more complex degree of involvement of farmers, as they are engaged in decision-making in earlier and more fundamental stages of the variety development chain; PPB therefore has a higher empowerment effect than PVS (Witcombe, 2005). Before proceeding, it is important to note that farmers’ interests in the outcomes of PPB or PVS rarely end with the evaluation of improved materials. Farmers’ ability to certify or multiply and distribute seed is directly affected in many countries by legal regulations and standard-setting bodies. PPB and PVS therefore can and should raise farmers’ awareness of those regulatory frameworks and, where possible, involve farmers in efforts to influence the modification of those frameworks if they limit farmers’ ability to maximize the benefits of exploiting the materials they participated in improving.


Why the need for participatory crop improvement?
In conventional plant breeding, the development of varieties is the responsibility of plant breeders; farmers are recipients of varieties only when they have been released by scientists and included into the official recommendation list of the agricultural extension system. The theory behind main reason for involving farmers is simple: farmers know best what they need and they make the ultimate decision on what is adopted. To ignore local knowledge and farmers preferences is very inefficient and modern cultivars are often rejected by farmers because of traits that have not been considered in the breeding process (LARC, 1995; Chemjong et al., 1995).


Contribution of PPB to farmers’ rights
PPB contributes to farmers’ rights in several ways. PPB provides farmers with the opportunity to influence the development of technologies in ways that are informed by their specific needs, agro-ecological environments and cultural preferences. In doing so, PPB increases considerably the likelihood that the final products will be well suited to the conditions of the farmers concerned and therefore contribute better to their livelihoods. Commercial conventional breeding tends to focus on producing varieties for resource rich farmers that are adapted to a wide array of non-extreme conditions with high production potential. These varieties may not perform well on land subject to extreme conditions, which is often the only land that resource poor farmers can afford (Gyawali et al., 2006). Non-commercial conventional breeders, such as state-owned extension stations and breeding/research institutions, commonly lack the resources needed to address the needs of resource poor farmers, including not having access to conditions comparable to those found on farmers’ fields and lack of knowledge of the specific needs of farmers, and hence may not be able to cater for farmers who produce under extreme conditions. The participatory approach is not only reflected in the different background of the people involved in the selection/breeding process, but also in the location of the (field) research. Test fields are located within the community and new varieties can thus adapt to real production conditions. The genetic material that is used is chosen together with farmers and has often already been cultivated in the community; this is another way to maximize adaptation to the local agro-ecological circumstances. PPB also involves providing farmers with access to genetic diversity as a basis for their innovation. In addition, as the participatory process progresses, the farmers develop an increasing sense of ownership as the lines they select in one breeding cycle are used to make crosses and start a new cycle. Such empowerment can lead to a cascading series of local innovations. Although PPB is particularly appropriate for developing products for use in more extreme environments, it is equally successful in improving materials for use in high-yield production potential environments. PPB provides farmers with the opportunity to influence decision-making about where financial resources for research, and agricultural extension services, should be dedicated. PPB makes use of the traditional knowledge of the farmers involved. It thereby elevates the profile of that knowledge and of the holders of it, creating incentives to continue using and developing it. While it is still not a very widespread practice, PPB can be structured to provide opportunities for women to participate. There are programmes in Eritrea, Jordan and Yemen where women’s participation has been encouraged through the presence of women among the scientific staff. In China, in many regions where men have migrated from farms to towns and cities, women are playing key roles in PPB processes. Participatory processes bring farmers into contact with professional breeders, thereby raising the farmers’ awareness of what science can offer them. This has an empowerment effect which is evident in the enhanced quality of farmers’ participation over time; they become true research partners. PPB involves mixing local as well as introduced improved materials. As such, PPB facilitates knowledge and technology transfer and capacity strengthening, both for the farmers and the formal-sector breeders with whom they come into contact. In more broadly conceived PPB projects, which focus on downstream use and dissemination of PPB products, farmers are involved not only in breeding activities, but also in the registration of the variety produced, its maintenance, seed multiplication and distribution, and, as appropriate, commercialization. In some cases, it may be worthwhile to explore obtaining formal rights over such materials, whereby the farmers involved, their associations or communities could be recognized, based on their contributions through the participatory process. PPB can strengthen farmer seed systems, defined as the ways in which farmers produce, select, save and acquire seeds (Weltzien and Vom Brocke, 2001). A healthy seed system includes four important characteristics: (i) it maintains a germplasm base that provides diversity, flexibility and a base for selection; (ii) it produces quality seed for production (free of seed-borne disease, with high germination rates and vigorous); (iii) it ensures seed availability and distribution (seed sources, social networks, markets); and (iv) knowledge and information about the seed are available and shared (growing methods, utilization, knowledge of new materials) (Hodgkin and Jarvis 2004). PPB, properly executed, should contribute to each of these four components.


Additional advantages of PPB
By focusing on the development of varieties that are suited to particularly extreme agro-ecological niches, PPB supports the development and maintenance of a more genetically diverse portfolio of varieties. The success of PPB should therefore be measured more by the number of new varieties produced and used in those niche environments (and the improvements they contribute to farmers’ livelihoods) than by the total number of hectares sown globally to a particular variety. In Syria, where a PPB programme on barley started in 2000, 25 varieties have so far been selected, named and multiplied; each of them occupies between 5 and 25,000 hectares. Similarly, six barley varieties have been named and multiplied for their adaptation to the north-west coast of Egypt, and three varieties of barley and one of lentil are being multiplied by farmers in Eritrea. Other successful examples can be found in countries as diverse as China (maize), Nepal (rice and maize), Mali (millet), Cuba (maize, beans, rice, cassava and tomatoes) and Honduras (maize and beans). PPB provides a forum for building participants’ knowledge and skill in genetic resources conservation and empowers rural institutions and farmers in community-based crop improvement and biodiversity enhancement (Sperling et al., 2001; Sthapit and Rao, 2007). PPB can also be less costly to conduct than traditional breeding, due to potential savings on field testing sites, lower overhead costs and the shortening of the research period required for producing useful materials. One study reports that “on-farm crop variety evaluations revealed a cost of US$0.50 per recorded data point for participatory trials, compared with $0.80 for conventional trials” (Toomey, 1999). A recent cost-benefit analysis of participatory and conventional plant breeding conducted in Syria shows that the benefit/cost ratio of PPB is 2.6 times higher than that of conventional plant breeding (Mustafa et al., 2006). PPB can speed up varietal development and release, as noted by the World Bank. It is too early to say that such savings will always be possible, and more economic analysis would be helpful. Early indicators, however, provide reasons to be optimistic. Furthermore, it has been demonstrated that PPB products are not necessarily local in nature and can be international public goods, as demonstrated in the case of rice varieties developed in Nepal that are used in India and Bangladesh (Joshi et al. 2006; Joshi, Sthapit and Witcombe, 2007). Similarly, PPB varieties developed in Syria are also being grown in Iraq. There are numerous such examples. However, the most important global public goods resulting from PPB work so far are the methods and models for engaging farmers in breeding and research. These can be replicated and adapted for use with many different crops anywhere in the world.


The global status of PPB
Despite the advantages of PPB listed above, its long-term adoption in national and international plant breeding programmes has been slow. Short-term projects are more common. The Working Group on Participatory Plant Breeding (PPBWG), established in 1996 under the framework of the Consultative Group on International Agricultural Research (CGIAR) System-wide Program on Participatory Research and Gender Analysis, estimates that nearly 100 participatory breeding projects are currently being implemented worldwide. The International Center for Agricultural Research in the Dry Areas (ICARDA), for example, has ongoing PPB programmes in Algeria, Egypt, Eritrea, Iran, Jordan and Syria on barley, wheat, lentil, chickpea and faba bean. International Center for Tropical Agriculture (CIAT), the International Potato Center (CIP), the International Maize and Wheat Improvement Center (CIMMYT), the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and the International Rice Research Institute (IRRI) have also been involved in PPB projects. Still more initiatives are led by national agricultural research organizations and non government organizations, with many examples in Asia, Africa and Latin America (Vernooy, 2003).

Making the most of PPB for farmers: policy challenges and options, Policies and laws have a significant impact on the degree to which farmers are able to take full advantage of their involvement in PPB. For example, in some countries only scientists with an MSc or higher degree can apply to register a variety in the national register; registration is a prerequisite for the commercialization of the variety (and, in some countries, to even being able to make it available for free). This would preclude most farmers engaged in PPB activities from being able to register new varieties in their own names or that of their association or community. Furthermore, in a growing number of countries, varieties must conform to relatively strict standards of distinctness, stability and, especially, uniformity before they can be registered. Some PPB products—while potentially very useful for farmers—will not be able to satisfy these criteria. The policy objective behind the introduction of these standards is, of course, perfectly justifiable— consumer safety, particularly that of farmers buying seed. But if applied too strictly the standards may limit the ability of farmers to exploit commercial markets for their PPB products. It is also important to recognize that PPB, like conventional breeding, is flexible and can be used to produce varieties with different genetic structures, including pure lines and hybrids. In some countries, new research is now underway to identify appropriate policy and legal options to support PPB, in order to keep the diversity-friendly advantages of the approach (Ceccarelli and Grando 2007). Similarly, seed laws setting standards for the conditions under which registered seed must be maintained and multiplied can also have the effect of squeezing out farmers who participate in PPB. Although the public policy goals of such regulations are important, the challenge is to find a middle ground where the benefits to farmers and diversity maintenance goals are taken actively into consideration. It is important to recall that, in many developing countries, up to 90% of seed planted is produced by farmers, not by seed companies. Release of PPB products into existing informal seed systems is a potentially very efficient, low transaction-cost approach. Accommodations of this nature—e.g. relaxing uniformity standards for landraces or PPB products in national laws—are not without precedent. The testing guidelines for the Nepalese Seed Law have been amended to allow lower levels of uniformity for some products. The draft European Community Directive on Conservation Varieties also allows for the registration of so-called conservation varieties pursuant to relaxed standards and marketing of those varieties.


Types of Participation
Biggs (1989) categorised farmer participation in agricultural research into four categories: contractual, consultative, collaborative and collegial. White (1996) also used four categories which corresponded precisely to those of Biggs (1989) but with different names: normal participation, consultative participation, action oriented participation, decision making/design participation. There is no exact analogy for the typology of participation described by Arnstein (1969) in agricultural research. The closest is that of Pretty (1995) who defined seven classes of participation — passive participation, participation resulting in information transfer, participation by consultation, participation for material incentives, functional participation, interactive participation and self-mobilisation. Some of these match those in the ladder of citizen participation while others match those of Biggs (1989). Many workers in participatory research have found the definitions of Biggs (1989) the most useful. Biggs did not give different values to the participation types but pointed out that mangers need to be clear about which mode is appropriate at a specific time and need to create a working environment that promotes that mode. The appropriateness of any mode of participation depends on its goal. If it is the empowerment of farmers, then the greater the involvement of farmers the better the ultimate goal is collegial participation where farmers run their own PPB programmes in a way that should be self-sustaining. However, if the purpose is functional, then contractual or consultative forms of participation may be the most effective.


Contractual participation
Contractual participation have used to obtain land for many of the breeding procedures e.g., generation advance, small plot multiplication, and progeny trials. Farmers were given compensation for the area used by the project. This was assumed to be equivalent to the yield the farmer would normally have obtained from that area, and yield per unit area was assessed by a crop cut in a farmers’ field that was mutually agreed between the farmer and project staff. This contractual arrangement is essential when, an NGO has no access to land of its own. A second example of contractual participation is when we have compensated farmers for labour involved in individual plant selection in very large plots. This occurs at the early stage of modified bulk breeding when a very large F4 or F5 population is divided into sub-bulks according to phenotype. This contractual participation creates the genetic material for subsequent collaborative participation.


Collaborative participation
The normal mode of participation for all of the PVS trials is collaborative, where farmers conduct experiments on their own fields. In PPB, participation is also collaborative when farmers grow in their own fields the bulks they have been given and select plants from them. Collaborative PPB appears to be a very cost effective since selection can be replicated across environments and across the individuals who carry out the selection.


Consultative participation
Nonetheless, consultative participation in PPB has already proven its value in goal setting, in the selection among lines in the field, and in the organoleptic testing of varieties. It is clear that successful PPB can be done entirely in a consultative mode, although collaborative PVS is required to test the genotypes produced by the PPB.


Collegial participation
Although empowerment is not a primary aim of functional approach, collegial participation is evolving on its own. In the terai (Nepal), farmers collaborating the most are forming a “breeders club” and wish to put their collaboration with NGO on a more formal basis. They explain that, at present, agreements are on a personal level, but an institution is a more assured long-term partner than individuals. The participation will be increasingly collegial because they intend the club to have a strong role in agenda setting.


Activities included in PPB
• Identifying breeding objectives
• Generating genetic variability (including the provision of plants to be included in breeding program)
• Selecting within variable populations to develop experimental varieties
• Evaluating experimental varieties (PVS - participatory variety selection)
• Variety release
• Popularization (diffusion of information about new variety and and how it is managed)
• Seed production


Participatory Plant breeding adding value in Nepal
One approach to benefit sharing involves creating reward and support systems that allow farmers to profit from the contributions they make to the global genetic pool. This can be done by adding value to the crops they grow, which again can contribute to improved livelihoods and increased income. As will be seen from this example from Nepal, this can be possible when farmers and scientists collaborate in participatory plant breeding (PPB).

In recent years Nepal has been giving greater priority and attention to the conservation of its rich biodiversity. Conservation efforts have largely been targeted at the country's many protected forest areas, national parks and reserves, but agricultural biodiversity is now gradually being recognized as an important component of the national biodiversity and worthy of conservation efforts. The value of agricultural biodiversity for Nepalese farmers and thus the importance of conserving it have been further established by research and development initiatives undertaken in the last 10 years. Local Initiatives for Biodiversity, Research and Development (LI-BIRD), a civil society organization, has been a pioneer in promoting on-farm conservation of agricultural biodiversity in Nepal since 1997. Working with several international and national partners, among them Biodiversity International, Nepal Agricultural Research Council (NARC), the Department of Agriculture and community-based organizations, LI-BIRD has identified various good practices for community-based on-farm conservation of agricultural biodiversity.
Traditionally, farmers in Nepal have maintained a high degree of agricultural biodiversity on their farms and in their communities. More than 90% of their propagating material has come from their own production or farmer-to-farmer exchange. In addition to being vital to the maintenance of agricultural biodiversity the local seed-supply systems have been crucial for the food security of resource-poor farmers. But also in Nepal the agricultural production system has been affected by technological changes and greater integration into the market economy. This has resulted in a gradual loss of agricultural biodiversity and need to restore traditional knowledge and conserve biodiversity. LI-BIRD's experiences in Nepal show how strategies that provide farming communities with incentives to act together and that benefit farming households have been helpful in promoting on farm conservation of agricultural biodiversity. These strategies capitalize on the opportunities for conservation inherent in the utilization of genetic resources for meeting cultural and development needs - especially strategies based on social values, and strategies based on economic incentives. The former promote on-farm conservation of agricultural genetic resources by increasing their uses in the socio-cultural rituals; and by providing social recognition and awards. Strategies based on economic incentives involve conservation through value addition aiming for increased production, desired traits of economic value, together with increased marketing and thus a higher cash income.

LI-BIRD has been promoting approaches which support farmers and farming communities in taking the lead role in the conservation and utilization of agricultural biodiversity. These approaches are referred to as good practices for on-farm conservation of agricultural biodiversity and are collectively known as community-based biodiversity management. These approaches involve raising the understanding of local knowledge and practices on the cultivation and use of the community genetic resources, and building the capacities of local community-based organizations and farming communities to plan and implement conservation and utilization strategies. The measures employed include seed fairs, a community biodiversity register, a community biodiversity fund and a community seed bank.
In Nepal, rural and urban consumers generally prefer local plants and their products for their taste, as well as their associations with family tradition and cultural rituals. However, due to low productivity and low volume of production, marketing of many of the local plants is difficult and usually not profitable. On-farm conservation of such plants is therefore often endangered because fewer and fewer farmers grow them. LI-BIRD and several NGOs have been working with several farming communities to improve the perceived value of many under-utilized crops by adding value through processing and packaging, and then marketing them as quality food. Local crops are also promoted by using them to make non-traditional modern food, like Western-type bread, cakes, cookies, noodles, and so on in an attempt to attract young people. Because of these interventions, the production area of local crops like finger millet, anadi rice (a sticky rice with medicinal and cultural value), buckwheat and taro has been steadily increasing in the farming communities participating in the programme.

LI-BIRD's extensive experience in participatory plant breeding has successfully been used for on-farm conservation of local rice varieties. The basic principle of the conservation-oriented PPB is to add value to the local plant varieties by further developing traits with economic or socio-cultural value and conserving the genes of these varieties in the process. Jethobudho - an aromatic rice land-race of the Pokhara Valley - was enhanced though PPB and has now been formally registered by the national variety release authority. As a result, farmers and farming communities in the area now possess ownership rights to Pokhareli jethobudho, the enhanced Jethobudho variety. Grassroots-based breeding programmes of this type have also promoted farmers' innovation in local crop development.
PPB has been used to combine the conservation of plant genetic resources with development goals. Both farmers and scientists increasingly appreciate PPB as a viable strategy for combining conservation with development goals in farming communities.


History of Participatory PLB in Nepal
The Lumle Agricultural Research Centre in Nepal, along with other DFID-supported research efforts in Nepal, had a long history of using participatory methods and was instrumental in developing new methods for on-farm research (Pound et al., 1988). One example was the development of informal research and development (IRD) techniques (Joshi and Sthapit, 1990). The lessons on farmer participation that we describe are from two separate breeding programmes: one for high-altitude, partially irrigated rice, the other for low-altitude, irrigated rice grown in the terai of Nepal.

Rice in high-altitude
The adoption of rice varieties released for high altitudes had been disappointingly low. In search of a more effective approach, plant breeders of LARC decided in 1985 to work directly with farmers and involve them at an earlier stage of selection. At this time, LARC was collaborating with CAZS, University of Wales, who were also concerned with developing methods of participatory research (Joshi and Witcombe, 1996).

High-altitude rice breeding was initiated on-farm in two high-altitude villages of Chhomrong and Ghandruk. In Chhomrong existing rapport with farmers strengthened the case for participatory research. When the participatory plant breeding (PPB) programme started, LARC had an employee based in the village who knew all of the villagers well. Farmers were already involved in the LARC farming system research programme (land was rented from farmers, and were employed for cultivation of the crop). Farmers collaborated in the breeding research because it gave them access to new germplasm and they wanted a better rice variety.

Rice in the terai
Several years after the high-altitude rice breeding programme, LI-BIRD and CAZS commenced research on participatory varietal selection (PVS) and PPB for the high potential productions systems of the Nepal terai. Although the uptake of improved varieties was high, many of the varieties that farmers were growing were extremely old. It was wished to test if participatory crop improvement would be effective in a high potential production system, because the use of such methods in favourable environments had never been well documented and there were only rare mentions of this approach in less formal (grey) literature.

Detailed material and methods on the research in the terai are reported in Witcombe et al., (2001). In the literature, participatory varietal selection (PVS) was the first method used for varietal improvement. This was an amalgamation of the terminology of Sperling et al., (1993) who used 'varietal selection’ for the process of selecting varieties, and 'participatory selection' for the participatory approach to varietal selection.


Different Knowledge systems of Farmers and Plant Breeders
Farmers' knowledge-quality traits


Goal setting for quality
At the commencement of the PPB programme, scientists consult with groups from the communities. The participation of farmers is crucial in setting goals as they know better their needs in new varieties. The breeders do not have a breeding objective at the outset of the programme.


Selection criteria for quality
As the research progress, other quality traits emerge for which farmers have a better knowledge of their importance (Sthapit et al., 1996). Quality is not a fixed attribute but varies with locality, socio-economic class, and use. The participation of farmers provides insights into their local preferences. Hence, in high- altitude village’s expansion on cooking and appetite delay are important, while in the terai market price is paramount unless it is for the farmers’ own consumption when aroma and softness become important.

For example, Chaite rice (February sown) variety, Radha 32 very is high yielding (producing nearly 25% more than the most widely grown variety CH 45), and has multiple disease and pest resistance including resistance to brown plant hopper an increasingly important insect pest of rice in the terai. Breeders were confident it would be a highly successful variety particularly as it initially spread rapidly from farmer to farmer (Witcombe et al., 2001). However, farmers soon rejected it because it had very poor cooking and eating quality and a low milling recovery i.e., a low ratio of grain to the other products of milling such as chaff. PNR 381 rice variety was widely adopted after the first year of PVS. In the next year, it was not grown in about 60% of the villages because of its poor post-harvest traits (Joshi, 2001).

Gauchan et al., (2001) found that the market provides incentives and disincentives for the farmer’s choice of variety through price signals, market margins and market channels. Despite this, hardly any effort is made in the formal varietal testing system to assess the potential market for new rice varieties.


Farmer’s knowledge - threshing and shattering
Easy threshing is a desirable trait in Nepal because threshing is not mechanised. Easy-to-thresh varieties require fewer beats of the rice bundle against the ground or a hard surface to remove the grain from the panicles. This increases labour efficiency and reduces drudgery. Difficult-to-thresh varieties obviously require more beatings to separate grains completely from the rice bundle. Less obviously, farmers consider plant height an important contributing factor to easy threshing. For simple mechanical reasons, if the straw is short, more beatings are required to remove the grains (Joshi et al., 1997). Easy threshing is a particularly important selection criteria in Chaite rice production because threshing, which coincides with the monsoon rains, need to be finished quickly. Farmers rejected IR13155 variety of Chaite rice in the lower hills and the terai because it was hard to thresh (Joshi, 2001). Farmers at Chhomrong village rejected Machhapuchhre-2 because it was extremely difficult to thresh (Sthapit et al., 1996).
Farmers can evaluate better than breeders both threshing and shattering. Screening requires a practical test that farmers apply as a matter of course but there is no easy way to reliably screen for shattering and threshing in formal breeding trials. Hence, in formal breeding programmes, ease of threshing is usually not assessed, and hard-to-thresh varieties would only be noticed if they could not be threshed by the mechanical thresher that is typically employed.


Farmer’s knowledge - other traits
Farmers have a more intimate knowledge of the crop. They grow and manage the crop themselves from sowing through to harvesting and threshing, so they have ample opportunity to study all of the crop traits and its adaptation. Unlike breeders, farmers are involved in all post-harvest processes. Farmers in the terai did not prefer Pusa Basmati-1 rice variety, although it fetched a premium price in the market, because they found that its long awns made milling more difficult. Farmers evaluate complex traits that need close observation of the crop over time, such as proneness to shattering. Chhomrong farmers considered the stay green character of M-3, as an indicator of good straw quality but this is rarely, or never, used as a selection criterion by rice breeders. It is undoubted that farmers minutely observe traits that concern them.


Farmers’ choice of testing sites
Farmers' strategies of selecting trial site drastically differ from that of scientists. Breeders often optimise the growing environment by applying purchased inputs, irrespective of the farmers' practices, since this tends to even out within-field environmental heterogeneity and allows a better expression of the genetic potential for yield. Farmers adopt what appears to be a very careful strategy for choosing the testing site of new crop varieties by exploiting their detailed knowledge of the niches within their village and fields. Farmers are simply adopting a risk-avoidance strategy if the new variety fails in the hostile environment, it does not matter because in the same area the local variety also produces little. Perhaps it is a more sophisticated if the farmer is going to adopt a new variety, it should perform well in poor, as well as in more favourable, environments. The farmers may feel that a good performance in the poor environment is an indication of excellent performance in a favourable one.


Farmers identify niches
Farmers discover subtle ecological preferences. In the terai, Swarna rice variety was initially tested in participatory varietal selection (PVS) trials that included varying water regimes. Farmers found it to be best suited to long-standing water areas where usually Masuli yields much less because it lodges. The area under Swarna is rapidly increasing now farmers are convinced of its advantages over Masuli in this particular niche. Rice variety BG 1442 was identified by breeders for the Chaite season and farmers selected it in PVS trials for this season. However, farmers also multiplied it during the main season and because it performed well it is now becoming popular in both seasons. Kalinga III rice variety is very early maturing. In areas where farmers grow two rice crops a year, farmers grow Kalinga III in areas where they will raise the nursery beds for the main season rice. This is possible because Kalinga III matures nearly 15 days earlier than the widely grown Chaite rice variety CH 45.


Breeders’ knowledge - in the process of plant breeding
Farmers could observe and understand the different steps in the breeding process and scientists assisted farmers in taking care to maintain varietal purity by avoiding the mixing of different genetic materials. Scientists are able to explain to farmers the concepts of genetic segregation in the progeny of crosses and the concept of heritability. Scientists are able to share with them the need for selection and selfing over several generations in order to develop a population of plants that are more uniform and stable. This help them to make distinction between advanced generation lines that need little time and effort to make uniform, and F2 or F3 generations that require many years and considerable effort. They also learn, when they see the progeny of several crosses involving Chhomrong dhan, the way that siblings from the same parents can differ. For example, farmers observed that a few lines had a grain type and colour very similar to Chhomrong dhan, but were very different to it for other morphological traits. In contrast, some varieties that were morphologically very similar to Chhomrong dhan had white grains.
In general, scientists have a greater knowledge of new germplasm. Breeders also have more experience in hybridisation so they make the crosses and the segregating progenies are given to farmers. Hence farmers are not trained to emasculate florets and to pollinate, which of course would be done if the goal is empowerment rather than to simply increase efficiency. Breeders and pathologists have a better knowledge of plant pathology, and understand the symptoms of micro-nutrient deficiency. They assisted the farmers in diagnosing the causes of symptoms and advised on those that could be safely ignored, and those that are important. For example, sheath brown rot is an important disease at higher altitudes and farmers are told; how to recognise it; that selection against the disease would be effective; and that taller plants with better panicle exsertion will give both improved disease resistance and better chilling tolerance (Sthapit et al., 1995).
Breeders knowledge - after varieties have been produced
Breeders enter M-3 into the formal trials and test it across agro-climatic zones. This is possible for formal-sector scientists but, because of a lack of knowledge and access, it would be unlikely for a farmer to enter his or her varieties in multilocational trials. Knowledge of the theory of multilocational testing that accounts for variations in climate, pathogens and pests across locations allows scientists to select for widely adapted varieties that perform well across-locations. Multilocational testing is combined with PPB to not only provide varieties that are specifically adapted to the needs of the participating farmers but to identify those that meet the needs of farmers far from the original area of the PPB programme.

The importance of the informal seed supply system (farmer-to-farmer networks) is very great in Nepal because formal extension and seed supply systems are poorly developed. In part this is because of the difficulty of transporting seed over the mountainous terrain. It often has to be carried on foot. Baniya et al., (2000) reported that more than 97% of rice area is planted with farmers own seed (of both landraces and modern varieties) and 85% of farmers regularly obtain rice seed in social seed networks by exchange, gift, or purchase. However, despite its importance, the speed of farmer-to-farmer spread is relatively slow and it tends to be limited within geographical areas.


Novel Interpretation of Plant Breeding Theory
Much of plant breeding theory is uncontroversial and not open to very different interpretations. The theories of heritability, selection differential and selection response are straight forward and do not differ theoretically between conventional and participatory methods. In practice, if heritabilities are expected to differ between farmers’ fields and the research station it might present a case for or against PPB. In Nepal, at least in the target regions in high altitude rice and low-altitude irrigated rice in the terai, there is no significant difference between the heritabilities found in farmer’s fields and on research stations. Sthapit et al., (1995) report on the heritabilities of field resistance to sheath brown rot disease in these crosses of high-altitude rice in Nepal. Sthapit (1994) reported on heritabilities for many characters in three crosses and in four locations. There was no significant difference in the heritabilities between the research station and the other locations where the crop was grown on farmers’ fields. In the high potential production system of the terai farmers’ fields tend to be extremely uniform and there is no reason to suppose that the uniform farmers’ fields will be more heterogeneous than research-stations fields.
Selection theory; i.e., what selection units to use, and the intensities of selection that can be used in each generation are open to greater interpretation. For example, in outbreeding crops there is much empirical evidence on whether half-sib, full-sib, S1 or S2 families should be employed as the selection units or, indeed, whether mass selection is more cost-effective (e.g., for maize, Hallauer and Miranda, 1998). In inbreeding crops there is an equally voluminous literature, but no comprehensive review, on the efficiency of various breeding methods — e.g., pedigree, bulk-pedigree, bulk, modified bulk or single seed descent. Nonetheless, in practice most methods, if executed in reasonable accordance with theory, will result in a genetic gain and plant breeders have adopted methods that best fit the programme’s human resources and infrastructure and its level and duration of funding.

It is clear from the economics of plant breeding theory that genetic improvement should be for traits of economic importance (e.g., Simmonds, 1979). However, formal trial systems emphasize yield too much, measure a limited number of traits, and rarely if ever, employe a system of trade-offs between those traits (Witcombe et al., 1998). Hence, a major impetus for participatory research is multi-trait evaluation by those that make the final decision on whether a variety is adopted or not the farmers. Farmers evaluate in focus group discussions many traits and trade them off e.g., by accepting lower yields for earlier maturity or higher quality. The results have somewhat different implications for PVS and PPB:
• In PPB, once the important traits are identified and the trade-offs among them are known, collaborative participation can be avoided. All the farmer-identified traits can be evaluated by scientists in their on-station breeding trials or in laboratory tests of harvested grain. A selection index can be created using the relative weights that farmers’ give to traits. Even if the selection index is less than optimal it increases the general appropriateness of the material that is bred by the programme to meet farmers’ needs.
• In PVS, the case for the continuing involvement of farmers is stronger. The ultimate purpose of any multi-trait evaluation is to predict whether farmers will adopt a variety, and PVS provide a simpler and more direct approach than using a selection index in multilocational trials. PVS also perfectly accounts for any changes in farmers’ preferences over time with, e.g., changes in market price and fodder availability and for differences between socio-economic groups and localities. Using selection indices in multilocational trials is more complex than using PVS. Since PVS is required to verify the selection indices employed it is a reasonable strategy to also use PVS to test varieties.
Selection in the target environment should be more effective and lead to better adapted, albeit more specifically adapted, varieties (e.g., Simmonds, 1984; Ceccarelli and Grando, 1989; Altin, and Frey, 1989). Most methods of analysis that have been proposed to account for G x E (e.g., reviewed by DeLacey et al., 1996) concentrate on analysis of yield and they have become increasingly complex, increasingly less intuitive and all are completely non-participatory. Only the methods of Binswanger and Barah (1980) have ever attempted to take farmers’ perceptions, in this case of risk aversion, into account.
Unfortunately, no matter how sophisticated the method of analysis, the G x E that occurs in the analysis is dependent on the environments of the test sites and these are often unrepresentative of target environments. Frequently the test sites fail to represent the diversity and the mean of the environments in farmers’ fields (Packwood et al., 1998). Increasing testing with farmers, and reducing the emphasis on formal multilocational trials, is one of the best ways of improving the appropriateness of test sites. There is also a gain in exploiting specific adaptation to a target environment, although this advantage may be outweighed by the reduction in the scale of adaptation. Unfortunately, theory cannot predict an expected degree of specific adaptation for the genotypes that emerge from a breeding programme, i.e., for the new genotypes what will be the slopes of the genotype regressions against an environmental index (Finlay and Wilkinson, 1963) compared to those of control varieties. What little empirical evidence exists is insufficient in quality or quantity to give a reliable prediction of the outcome.
Once the decision has been made to decentralise to farmers fields, to have the benefits of multi-trait evaluation and to exploit specific adaptation, practical issues intervene. Moving a breeding programme to farmers’ fields meant that traditional plant breeding practices that rely on the generation of a very large number of test units from many crosses were impracticable. Hence, we should use many fewer crosses and a much larger population size for each cross than is common in conventional centralised plant breeding. Even though it is contrary to common practice this is supported strongly by theory:
• “Choice of germplasm to be used in a practical breeding programme may be the most critical decision facing the breeder, although the choice of breeding method and system of evaluation can be equally important. Goodman (1985) further pointed out that choice of materials is perhaps the most critical to the success of the programme. When one reviews the literature or reflects on plant breeding courses, however, far more time is devoted to breeding methods than to the selection of useful germplasm” (Maunder, 1992).
• “This [the very high number of possible recombinants] suggests that populations should be as large as practicable in early generations, and that their management should be directed towards obtaining, in later generations, manageable numbers of promising pure lines for intensive evaluation of agricultural worth. Such evaluations, it is hoped, will reveal at least one pure line deemed good enough to be released to farmers. If not, at least some pure lines may be found that the breeder considers to be worthy of use as parents in additional cycles of hybridisation" (Allard , 1999).
It is found that participatory methods assist greatly in reducing the number of crosses by reliably identifying superior parents. Witcombe and Virk (2001) have examined in more detail the strategy of low-cross-number, high-population size breeding.


How has our understanding of Plant Breeding theory and practice changed?
Why most breeders don’t do this?
The question arises as to why most breeders do not adopt a low-cross-number strategy. One reason is that they are not in the situation of starting a breeding programme where the predominant variety has been cultivated by farmers for many years and where there is no ongoing breeding programme. In this situation the choice of parents is simple. When there is an ongoing breeding programme, and farmers are constantly adopting new varieties, then there are many more parents to choose from.

Another reason is that many breeders work in a competitive environment, where to be commercially successful it is not enough to breed a variety better than the one farmers are growing. In commercial environments breeders have to produce the highest-yielding cultivar in official trials i.e. one that is better than those of all competitors. To do this breeders attempt to use the best possible parents. Given a genetic gain of 1-2 % per annum by plant breeding (e.g., Byerlee and Heisey, 1990), the most recently produced varieties are most likely to be highest yielding and, hence, on performance per se are the best possible parents. Unfortunately, these are the varieties about which least is known there is a tradeoff between the increased probability that a new variety is the highest yielding genotype and the quantity of information on performance per se. Hence, to ensure that the best possible parents are included many crosses are made among the newest material e.g., a diallel cross or partial diallel cross among the entries of an advanced yield trial. Of course, there is a cost to attempting to include the best parent as more crosses mean that individual population sizes of each cross have to be smaller. Also it is certain that many of the parents would not have been used if more were known about their performance.


Bulk-population breeding methods
Bulk population methods of plant breeding are more suited to participatory approaches (Witcombe et al., 2001). The theory supports this approach despite many breeding programmes relying on pedigree breeding. Bulk methods preserve variability in the early generations when plant-to-plant heritability is lower and selection begins when a high level of homozygosity has been achieved. The literature on such methods, particularly the adoption of single seed descent methods, indicates that these are efficient and successful (e.g., Fahim et al., 1998) in rice.

Practice: manifold changes in approach
Plant breeding practice has changed dramatically with the adoption of PPB approaches. These have resulted in:
• the selecting parents for PPB on the basis of performance and acceptability in PVS trials;
• the selection of target traits on the basis of consultation with farmers (this has changed the selection criteria used in the breeding programme);
• a reduction in the number of crosses and an increase in the population size of the crosses;
• a reduction in the use of pedigree breeding and an increased emphasis on bulk population methods;
• an increased emphasis on consultative screening for post-harvest quality traits with farmers (this cost-effectively reduces the number of entries to be tested by PVS, and ensures that more widely tested entries are not rejected because of poor grain quality); and
• the testing of varieties first with farmers rather than in formal multilocational trials; the most farmer-accepted lines are later entered in these trials.


On-farm conservation
PPB is a strategy to strengthen the process of on-farm conservation by encouraging farmers to continue to select and manage local crop populations, and manage seed supply system through informal networks. Diverse farmer preferences, agroecological niches and farming systems will conserve a reservoir of genetic diversity on-farm. This reservoir can be considered as valuable pre-breeding germplasm.
Both PVS and PPB may have a negative impacts on the diversity of landraces, because both methods are intended to change the local crop population structure to make it more productive and useful to farmers. This is likely because the variable, segregating materials used in PPB are derived from landraces already in the local farming system.


Future Directions
It is concluded by discussing that how the results of PPB work so far will effect its future directions. One of the major results of the work has been a realisation as to how important it is to involve farmers in setting breeding goals. In more recent research, goal setting has been the result of collaborative research – PVS – where farmers have evaluated new varieties in their own fields. One example is variety Pusa Basmati-1 for which farmers identified deficiencies. A breeding programme was established with the objective of removing the two they considered most important the presence of apical awns which made milling difficult and dwarf plant height that reduced straw yield. Farmers consider straw an important product as they feed it to their livestock. The progeny of natural out crossing were selected from a very large irradiated population of Pusa Basmati-1 that were not only awnless and taller but had improved grain quality. Another example was the initial widespread acceptance of Radha 32 for its high yield and its equally widespread, subsequent rejection because of its very poor cooking quality. Radha 32 used as a parent and crossed it to Kalinga III, that has better grain quality, with the breeding objective of producing lines that a yield at least as much as Radha 32 but that do not have its poor grain quality.
Use methods that preserve a greater genetic diversity, such as bulk population breeding and equal seed descent. Equal seed descent can, using few resources, preserve a reservoir of genetic diversity in a heterogeneous, but homozygous, population of plants. It is simple to derive pure lines from such a population and then evaluate them jointly with farmers in consultative participation.
PVS able to give quick results, whereas the first PPB products take more than four years to produce, and even then their testing is constrained by seed quantities. What is becoming clear, and leading the direction of much of current PPB programme, is that PPB is the filling in a sandwich between two slices of PVS. The first slice identifies both potential parents and breeding goals, and the second immediately identifies those varieties produced by PPB that are acceptable to farmers.


Conclusions
PPB promotes farmers’ right. There cannot be sustainable use without farmers’ rights; farmers’ rights are meaningless outside the context of sustainable use. Governments should encourage and facilitate the adoption of participatory approaches by public and private sector institutions involved in breeding. To this end, among other things, they should adjust their regulatory frameworks, particularly concerning variety registration and maintenance and seed production and marketing, to ensure that farmers are able to gain maximum benefit from PPB programmes. To achieve wide-scale mainstreaming of PPB, technical assistance for national (and international) programmes will be necessary. Some such assistance is already available through the Food and Agriculture Organization of the United Nations (FAO), the CGIAR centres and civil society organizations with experience in the field. The Global Partnership Initiative for Plant Breeding Capacity Building (GIPB) will also include substantial emphasis on building capacity for PPB. The more that the work of all of these initiatives can be considered and guided by the Governing Body of the Treaty, the better. In addition to providing a forum for the exchange of experiences on PPB, the Governing Body could provide guidance on how countries can support PPB. Equally importantly, the Governing Body could work to define indictors to measure the impact of PPB on both the sustainable use of PGRFA and farmers’ rights.


References
1.The CGIAR coordinated a side event entitled 'Options for National Implementation of Farmers' Rights: Participatory Plant Breeding and Participatory Variety Selection' on June 12, 2006, during the First Session of the Governing Body of the International Treaty. Speakers included Jan Engels (Bioversity International), Pratap Shrestha (LI-BIRD), Ruaraidh Sackville Hamilton (IRRI) and Rodomiro Ortiz (CIMMYT). This brief expands on themes addressed during that event.
Allard, R.W. (1999) Principles of plant breeding. Second edition. John Wiley and Sons, Inc., Toronto. pp.91.
Arnstein, S.R. (1969) Ladder of citizen participation. AIP Journal Vol. XXXXV No. 2: 216-224.
Atlin, G. N. and K.J. Frey (1989) Predicting the relative effectiveness of direct versus indirect selection for oat yields in three types of stress environments. Euphytica 44:137-142
Baniya, B.K., Subedi, A, Rana, R.B., Paudel, C.L., Khatiwada, S.P., Rijal, D.K .and Sthapit, B.R. (1999). Informal rice seed supply system and storage systems in mid-hills of Nepal. In A Scientific basis of in situ conservation of agrobiodiversity on:farm: Nepal’s contribution to the global project (B Sthapit, M Upadhaya and A Subedi editors) pp 79-91.
Biggs, S.D. (1989) Resource-poor farmer participation in research: a synthesis of experience from nine National Agricultural Research Systems. OFCOR Comparative Study Paper No. 3. The Hague: International Services for National Agricultural Research (ISNAR).
Biodiversity in Nepal: Lessons Learned. Nepal Agricultural Research Council, LIBIRD, Bioversity and IDRC. pp 164–173.
Biswanger, H.P. and B.C. Barah (1980) Yield risk, risk aversion, and genotype selection: conceptual issues and approaches. Research Bulletin No 3, ICRISAT, Andhra Pradesh 502 324, India.
Byerlee, D., and P.W. Heisey (1990) Wheat varietal diversification over time and space as factors in yield gains and rust resistance in the Punjab. In: Heisey P.W. (ed). Accelerating the transfer of wheat breeding gains in farmers: A study of the dynamics of varietal replacement in Pakistan, pp. 5-24. CIMMYT Research Report No. 1. International Centre for Wheat and Maize Improvement, Mexico, D.F.
Ceccarelli S and Grando S. 2007. Decentralized participatory plant breeding: an example of demand driven research. Euphytica 155:349–360
Ceccarelli, S. and S. Granado (1989) Efficiency of empirical selection under stress conditions in barley. Journal of Genetics and Breeding 43:25-31.
Chemjong, P.B., B.H. Baral, K.C. Thakuri, P.R. Neupane, R.K. Neupane, and M.P. Upadhaya (1995) The impact of Pakhribas Agricultural Centre research in the eastern hills of Nepal: Farmer adoption of nine agricultural technologies'. Dhankuta, Nepal: Pakhribas Agricultural Centre.
DeLacey, I.H., W.G.A. Ratnasiri and P.D.N. Mirzawan (1996) Retrospective analysis of historical data sets from multi-environmental trials–case studies. In: Cooper, M. and G.L. Hammer (eds) Plant adatation and crop improvement. pp. 269-290. CAB International/IRRI/ICRISAT.
Fahim, M., M.P. Dhanapala, D. Senadhira, and M.J. Lawrence (1998) Quantitative genetics of rice. II. A comparison of the efficiency of four breeding methods. Field Crops Research 55:257-266
Finlay, K.W. and G.N., Wilkinson (1963). The analysis of adaptation in a plant breeding programme. Aust J Agric Sci 14:742-754.
Gauchan, D., P. Chaudhary, B.R. Sthapit, M.P. Upadhaya, M. Smale, and D. Jarvis (2001). A participatory marketing system research approach to analysing market based incentives and disincentives: A case study of rice, central terai, Nepal. Paper presented at the national workshop of in situ conservation of agrobiodiversity on-farm: Nepal project. April 24-26, Lumle, Pokhara, Nepal.
Goodman, M.M. (1985) Exotic maize germplasm: status, prospects, and remedies. Iowa State Journal of Research 59: 497-527.
Gyawali S, Sthapit BR, Bhandari B, Shrestha P, Joshi BK, Mudwori A, Bajracharya J and Shrestha PK. 2006. Participatory Plant Breeding: A strategy of on-farm conservation and improvement of landraces. In: Sthapit BR and Gauchan D, editors. Proceedings of a national symposium, 18–19 July 2006, Kathmandu, Nepal. On-farm Management of Agricultural
Hallauer, A.R. and J.B. ,Miranda (1988). Quantitative Genetics in Maize Breeding. Second Edition. Iowa State University Press, Ames.
Hodgkin T and Jarvis D. 2004. Seed systems and the maintenance of diversity onfarm: Introductory remarks. In: Jarvis D, Sevilla- Panizo R, Chavez-Servia JS and Hodgkin T, editors. Seed systems and crop genetic diversity on-farm. Proceedings of a workshop, 16–20 Sept 2003, Pucallpa, Peru. IPGRI, Rome.
Joshi KD, Musa AM, Johansen C, Gyawali S, Harris D and Witcombe JR. 2006. Highly client oriented breeding, using local preferences and selection, produces widely adapted rice varieties. Field Crops Research 100:107–116
Joshi, A and J.R. Witcombe (1996) Farmer participatory crop improvement. II. Participatory varietal selection in India. Experimental Agriculture 32:461-477.
Joshi, K.D. (2001) Rice varietal diversity and participatory crop improvement in Nepal. PhD Thesis, University of Wales, Bangor, UK.
Joshi, K.D. and B.R. Sthapit (1990) Informal research and development (IRD): A new approach to research and extension. LARC Discussion Paper No. 90/4. Pokhara, Nepal: Lumle Agricultural Research Centre.
Joshi, K.D., M. Subedi, R.B. Rana, K.B. Kadayat and B.R. Sthapit (1997) Enhancing on-farm varietal diversity through participatory varietal selection: A case study for Chaite rice in Nepal. Experimental Agriculture 33:335-344.
LARC (1995) The adoption and diffusion and incremental benefits of fifteen technologies for crops, horticulture, livestock and forestry in the western hills of Nepal. LARC Occasional Paper 95/1. Lumle Agricultural Research Centre, Pokhara, Nepal.
Maunder, A.B. (1992) Identification of useful germplasm for practical plant breeding programs. In: Stalker, H.T. and Murphy, J.P. (eds) Plant breeding in the 1990s. Proceedings of the symposium on plant breeding in the 1990s. pp.147-169. CAB International.
Mustafa Y, Grando S and Ceccarelli S. 2006. Assessing the Benefits and Costs of Participatory and Conventional Barley Breeding Programs in Syria [unpublished].
Packwood, A.J., D.S. Virk and J.R. Witcombe (1998) Trial testing sites in all India co-ordinated projects—How well do they represent agro-ecological zones and farmers’ fields? In: Witcombe, J.R., D.S. Virk and J. Farrington (eds) Seeds of choice: Making the most of new varieties for small farmers. pp. 7-26. Oxford and IBH Publishing Co. New Delhi and Intermediate Technology Publications, London.
Pound, B.P. , Budhathoki, K and Joshi, B.R. (1988) Mountain agricultural technology development and diffusion: The Lumle model, Nepal. In: Jodha, N.S., M. Baskota and Tej Pratap (eds) Sustainable mountain agriculture: Perspectives and issues. pp. 711-736. ICIMOD, Oxford and IBH Publishing Company, Pvt. Ltd. India.
Pretty, J. (1995) The Living land: agriculture, food and community regeneration in rural Europe. Earthscan Publications Ltd, London.
Simmonds, N.W. (1979) Principles of crop improvement. Longman Group Limited.
Simmonds, N.W. (1984) Decentralised selection. Sugar Cane 6:8-10.
Sperling L, Ashby JA, Smith ME, Weltzien E and McGuire S. 2001. A framework for analyzing participatory plant breeding approaches and results. In: Elings A, Almekinders CJM and Stam P, editors. Breeding for lowinput conditions, and consequences for Participatory Plant Breeding. Euphytica 122 (3):439-450 (Special Issue).
Sperling, L., M.E. Loevinsohn and B. Ntabomvra (1993) Rethinking the farmers role in plant breeding: local bean experts and on-station selection in Rwanda. Experimental Agriculture 29:509-519.
Sthapit B and Rao R. 2007. “Grassroots breeding”: A way to optimise the use of local crop diversity for the well-being of people. Paper presented at the Tropengtag Conference, 9–11 October 2007, Witzenhausen, Germany.
Sthapit B, Joshi KD and Witcombe JR. 1996. Farmer participatory crop improvement. III: participatory plant breeding, a case study for rice in Nepal. Experimental Agriculture 32:479–496.
Sthapit, B.R. (1994) Genetics & Physiology of Chilling Tolerance in Nepalese Rice. Ph.D. Thesis, University of Wales, Bangor, UK.
Sthapit, B.R., K.D. Joshi and J.R. Witcombe (1996) Farmer participatory crop improvement. III. Participatory plant breeding, A case study for rice in Nepal. Experimental Agriculture, 32:479-496.
Sthapit, B.R., P.M. Pradhanang and J.R. Witcombe (1995) Inheritance and selection of field resistance to sheath brown rot disease in rice. Plant Disease 79:1140-1144.
Toomey G. 1999. Farmers as Researchers: The Rise of Participatory Plant Breeding. IDRC Online Archive. Available at: http://www.idrc. ca/ev_en.php?ID=5559&ID2=DO_TOPIC.
Vernooy R. 2003. Seeds that Give: Participatory Plant Breeding, IDRC Online Archive. Available at http://www.idrc.ca/ seeds/ev-30294-201-1-DO_TOPIC.html
Weltzien E and Vom Brocke K. 2001. Seed systems and their potential for innovation: conceptual framework for analysis. In: Sperling L, editor. Targeting seed aid and seed systems interventions; strengthening small farmer seed systems in East and Central Africa. Proceedings of a workshop held in Kampala, Uganda, 21–24 June 2000. CIAT, Kampala, Uganda
White, S.C. (1996) Depoliticizing development: the uses and abuses of participation. Development in practice Vol. 6, No. 1. Oxfam, UK and Ireland
Witcombe JR, Joshi KD, Gyawali S, Musa AM, Johansen C, Virkand DS and Sthapit BR. 2005. Participatory plant breeding is better described as highly clientoriented plant breeding. I. Four indicators of clientorientation in plant breeding. Experimental Agriculture 41:299–319
Witcombe JR. 2005. Participatory Varietal Selection and Participatory Plant Breeding: The last 10 Years. In: Gonsalves J, Becker T, Braun A, Campilan D, de Chavez H, Fajber E, Kapiriri M, Rivaca-Caminade J and Vernooy R, editors. Participatory research and development for sustainable agriculture and natural resource management: A sourcebook. Volume 1: understanding participatory research and development. CIP-Upward/ IDRC. Available online at http://www.idrc.ca/ en/ ev-73443-201-1-DO_TOPIC.html.
Witcombe, J.R, K.D. Joshi, R.B. Rana and D.S. Virk (2001) Increasing genetic diversity by participatory varietal selection in high potential production systems in Nepal and India. Euphytica (In press).
Witcombe, J.R. and D.S. Virk (2001) Number of crosses and population size for participatory and classical plant breeding. Euphytica (In press).
Witcombe, J.R., D.S. Virk and A.G.B. Raj (1998) Resource allocation and efficiency of the varietal testing system. In: Witcombe, J.R. D. S. Virk and J. Farrington (eds) Seeds of choice: Making the most of new varieties for small farmers. pp. 135-142. Oxford and IBH