RESEARCH WORK

Academic discussion about farmers' experiments a synthesis

F. Leitgeb, Elena Sanz, Susanne Kummer, Racheli Ninio y C.R. Vogl

Working Group Knowledge Systems and Innovations Division for Organic Farming. University of Natural Resources and Applied Life Sciences Vienna (BOKU). Gregor Mendel Strasse 33, A-1180 Viena

E-mail: christian.vogl@boku.ac.at

ABSTRACT

This paper reviews scientific literature dealing with farmers' experimentation and innovation. For this a search was conducted in databases, books and journals. Based on this information it can be stated that during the evolution of agricultural systems, farmers' experimentation and innovation have been an integrated part of the development of the worlds' agricultural systems. The capacity of farmers to respond and adapt to external and internal changes is the basis for agricultural evolution. The farmer is part of the system in which he/she is conducting experiments and has direct interest in improving the current situation. The motivation to start farmers' experiments arises from the perceived necessity or wish to find solutions for certain problems or just from the farmer's curiosity to try something. Farmers experiment and innovate with their own methods, which usually differ from scientific ones. The level of complexity ranges between very easy and very complex. The experiments are mainly conducted on the basis of locally available physical and biological resources. They can be classified into technical, economical, social and institutional ones.

Key words: Local knowledge, farmers' experiments, innovation

The Pastos y Forrajes journal is a high level professional and academic means for the presentation of scientific results, in most of the cases based on experiments. For scientists experimentation is closely related to classical designs, replications, controlled treatments, structured monitoring and multivariate analysis, often including mathematical models for the comparison of variables and their influence on the indicators.

But experimentation is not only an academic tool; it is also an everyday activity of non-academic actors, even of farmers, in their effort to adapt agricultural techniques and processes to new realities. Lately these experiments have received attention in several countries. Anthropologists, sociologists and agronomists have shown their importance for rural development, food security, health and economic income. Unlike other countries, Cuba has a policy of support to experimentation by producers and to participatory processes of innovation in the countryside; besides, their results receive high public acknowledgement.

Such experiments are part of agriculture since the soil was first cultivated and animals were first domesticated, i.e., since thousands of years ago they are the base for the development of agriculture and human culture. Testing new methods and technologies, as well as experimenting and innovating, have been integral and common elements in the daily life of the farmer (Haverkort, 1991; Scheuermeier, 1997; Sumberg and Okali, 1997; Bentley, 2006; Richards and Suazo, 2006). The development of agriculture is associated to frequent changes in the socioeconomic, socio-cultural, political and agroecological levels. The agricultural evolution is based on responding to these changes by means of the experiments carried out by farmers in their agricultural system (Mak, 2001). The experimenting process is necessary to adapt the production form under different conditions, which vary according to the agricultural and social system. They experiment based on their knowledge and experiences with the objective of improving their agricultural system (Rajasekaran, 1999).

Due to the importance of these experiments for the development of agriculture, it is necessary to understand experimentation in a more detailed way. The objective of this work is to structure the current scientific discussion on the topic and present the results, as a contribution to the efforts to combine farmers' experimentation with scientific experimentation, aiming at developing agriculture for the welfare of the people.

The paper is based on a structured and scientifically documented search on key words, combining farm, farmer, with experiment, innovation and adaptation. During 2006 a bibliographic review was carried out of the published written and electronic literature, such as electronic catalogs of libraries, journals (e.g. Science Direct, Kluwer) and databases (e.g. Agris). After analyzing all the data a proper base was created, classifying the information related to experimentation; the diverse topics were separated and summarized in each chapter.

The search in literature was made as initial step in the project "Organic farmers' experiments", supported by the Austrian Foundation for Science (FWF). This project also includes field research between June, 2007 and August, 2008, in Cuba, under contracts of the Universität für Bodenkultur of Viena (University for Natural Resources and Applied Life Sciences, Vienna), with two Cuban scientific institutions: the National Institute of Agricultural Sciences and the Experimental Station of Pastures and Forages "Indio Hatuey" (of the University of Matanzas "Camilo Cienfuegos").

Development

1. Definition of terms

The most common word of farmers when speaking about the topic is «testing». The first definition one finds of this term is: `to make exam and experiment of the qualities of someone or something' (RAE, 2001). Farmers define `testing' widely, often as synonym of `experimenting', for example, as the activity of introducing something totally or partially new to exploitation and evaluate the success or failure of this introduction (Quiroz, 1999).

`Experimenting' means making operations destined to discover, test or prove certain phenomena or scientific principles (RAE, 2001). When farmers use this word, it often means: `to observe deeply the results of a change induced by them in their farm and test these results with the opinion or statements of others'. It also means: `comparing something already known to something unknown' (Stolzenbach, 1999).

A word closely linked to the topic of experimentation is `innovation'. An innovation is an idea, a practice or an object that is perceived as new by an individual or another adoption unit. It is of little importance whether the idea is objectively new, measured in the passage of time since the first use or discovery. `Invention' is understood as a really new idea or technology, i.e., discovered or created for the first time (Rogers, 1995).

The experiments and innovations of farmers are different but complementary processes. Experiments contribute to the creation of new knowledge, previous condition for the development of an innovation (Rogers, 1995) or invention. Experimenting is a dynamic process in a certain period before the development of an innovation or invention. Experiment and innovation are part of every farmer's experimentation (Rogers, 1995; Hocdé, 1997). If the results of a certain experiment are not satisfactory, an innovation or an invention is not developed. In any case farmers increase their experience and local knowledge through experimentation (Bentley, 2006; Richards and Suazo, 2006).

2. Justification of farmers' experiments

Rural zones are characterized by their diversity of conditions, for which the needs of the people who live in rural areas are different according to the site and, consequently, it is not possible that one innovation is applicable for all farmers (Reece and Sumberg, 2003); each individual has to adapt the innovation to his/her conditions (Niemeijer, 1999; Sumberg, Okali and Reece, 2003) through an experimentation process. Their capacity to experiment and innovate is an important part of the development of agricultural systems; they must adjust their working way and their agricultural system to the changes of the surroundings (Bentley, 2006; Richards and Suazo, 2006).

Farmers continuously experiment and innovate to maintain and improve agricultural production, but the scientific community seldom pays attention to the methodology and results of those experiments (Haverkort, 1991; Bentley and Baker, 2005). Nevertheless, they experiment, with or without scientific support. Until now there are not enough methods for documenting or divulging such experiments (Bentley, 2006). The activities of research centers and multinational enterprises, frequently minimize the importance of farmers' experimentation and, in many cases, their capacity to experiment and innovate is underestimated. Scientific research is not always based on the reality of the farmer's life, nor takes into consideration his/her ideological concepts, although personal values can influence the acceptance and adoption of innovations. Frequently scientists and extension workers use the hierarchical model to transfer innovations, without considering adequately local concepts or the economic, socio-cultural, environmental and technical conditions of farmers (Bunch, 1991). With this transference model, known as top-down, problems emerged, because the innovations were not oriented to the needs of the people of rural areas. An example is the gender issue in agriculture, because the role of women did not always have adequate attention by scientific research. A consequence was that innovations were neither appropriate nor applicable by women (Haverkort, 1991).

In many countries small farmers represent a little favored social class, while most of the scientists usually belong to the highest social classes. This fact causes a difference between both social groups (Hagmann, Chuma and Murwira, 1997), that science tries to eliminate, but without detailed knowledge of the local conditions of small producers. Scientists' solutions require high expenses and external inputs, to save time and money; but in most cases farmers try to save money instead of spending it (Haverkort, 1991; Bentley, 2006).

Farmers have specific local knowledge about the environmental conditions and local problems, as well as experience that a researcher can not have (Sumberg and Okali, 1997). Understanding the farmers' experimentation process is the base for a beneficial cooperation and participatory creation of new knowledge. Formal science has to accept the innovating capacity of the farmer to cooperate and develop more appropriate technologies (Bentley and Baker, 2005).

3. Factors that influence the farmers' experiments

Among the factors that influence the willingness to experiment with new methods or technologies, are the external and internal ones. The external factors are linked indirectly to the farmer and include changes originated at political, institutional, social, cultural or economic level. Other external factors could be the biophysical environment and agroecological conditions (Mak, 2001; Padel, 2005).

The internal factors are directly related to the farmer; they are: age, sex, social network, work organization, production process or farmer's budget; other factors can be: size of useful agricultural surface or type of agricultural production (Mak, 2001; Padel, 2005).

3.1 Agroecological factors

Topography, such as slopes or other land characteristics, influence the willingness for experimenting. Slopes can favor experiments to stop erosion (GebreMichael, 2001). Climatic changes, such as severe drought, can force the farmer to test new irrigation methods (Niemeijer, 1999; Sumberg et al., 2003).

If the quantity of variables is high, such as the varieties of plants cultivated, animals or even agricultural machinery under exploitation the possibilities for experimenting increase. In diversified agroecological situations there are more possibilities for a producer to experiment (Sumberg and Okali, 1997).

3.2 Socioeconomic factors

There is a significant relationship between communication networks and innovation capacity. A well-developed social network favors the exchange of ideas and technologies. The social connections and relationships of a farmer mean more opportunities for obtaining information, technology, capital and advisory (Wu and Pretty, 2004). A social network of experimenting farmers facilitates informal research and has an impact on the quality and quantity of experiments (Hagmann et al., 1997).

Social pressure can have a negative influence on the people with extraordinary ideas different from those of the agricultural society. Some experimenting farmers can be criticized by others and considered disrespectful of the traditional culture (Reij and Waters-Bayer, 2001).

Institutional or political decisions can cause changes at the socioeconomic level, as well as mean opportunities or threats for farmers; those that cause precarious situations induce the need to experiment (Mak, 2001; Padel, 2005). In the case of the changes that affect production negatively, the farmers perceive in experimentation their only possibility to sustain their family and adapt their working ways to them (Taonda, Hien and Zango, 2001).

The economic situation of the experimenting farmer has a different impact according to the country and it can also affect positively or negatively (Sumberg and Okali, 1997; Miiro, Critchley, Wal and Lwakuba, 2001; Reij and Waters-Bayer, 2001; Wu and Pretty, 2004). If farmers perceive a bad financial situation they can feel encouraged to improve it. Likewise, an insufficient economic situation can prevent the farmer from experimenting, due to the high risk of losing money (Sumberg and Okali, 1997; Quiroz, 1999).

The farmers that work part-time and have other incomes beside agriculture face different conditions. Additional incomes can affect positively their willingness to experiment, as they have more money for agricultural investment (Sumberg and Okali, 1997; Nasr, Chahbani and Reij, 2001). On the other hand, working part-time in agriculture implies dedicating less time to the field. Probably part-time farmers feel less need to invest in the future of agricultural production. In the literature about the topic the examples of farmers who are dedicated only to agriculture stand out (Critchley, 2000; GebreMichael, 2001).

3.3 Socio-demographic factors

The experiments are carried out by farmers from all the socio-demographic groups, independently from age, gender, educational level, marital status or work situation (Hocdé, 1997; Sumberg and Okali, 1997; Critchley, 2000). Although some farmers are more actively involved in the experimentation processes and develop new methods and technologies or modify significantly the innovations introduced externally, it is difficult to identify the socio-demographic factor responsible for this attitude (Zigta and Waters-Bayer, 2001; Sumberg et al., 2003).

The experimenting farmers that develop innovations or inventions have some socio-demographic characteristics different from those that only adopt innovations created by others; the former have a higher educational level than the latter (Miiro et al., 2001). In addition, they generally maintain more cosmopolite relationships, travel frequently out of their towns and have other experimenting and innovating farmers in their social network. Nevertheless, there can be members within the local social network that do not accept the experimenting and innovating character (Rogers, 1995).

The farmers with large agricultural surfaces are usually less motivated to experiment than those with small plots (Hagmann et al., 1997; GebreMichael, 2001). According to other sources, there is no correlation between the willingness to experiment and the size of the population (Nielsen, 2001); it is even stated that experimenting farmers have large agricultural surfaces (Miiro et al., 2001).

Some authors state that experimenting farmers are between 30 and 50 years old (Nasr et al., 2001), but according to other sources they can be older (Miiro et al., 2001; GebreMichael, 2001). Older farmers are more experienced, which gives them advantages to experiment (Miiro et al., 2001). Young farmers, who recently started to work in agriculture, are still in a learning process, which leads them to the need of experimenting (Reij and Waters-Bayer, 2001).

Both genders carry out experiments, although there is a trend that sustains that men are more involved in the experimentation and innovation topic (Sumberg and Okali, 1997; Miiro et al., 2001). Three fourths of the experiments are performed by men. The question is whether it makes any sense to distinguish between men and women, because the decisions about important changes in the farm management, as well as the introduction of new technologies or methods require the agreement and support of the family. Some innovations carry such a risk that they require, at least, consultation with the family or even their active participation (Reij and Waters-Bayer, 2001).

There is a separation of tasks in a farmer family. Men work in the field, which explains that most of the experiments are carried out by them (Hocdé, 1997; Reij and Waters-Bayer, 2001; GebreMichael, 2001). Women have wide knowledge in topics such as seed storage, food elaboration, medicinal plants or marketing (Gupta, 1996); more experimenting women are found in such topics (Hocdé, 1997; Reij and Waters-Bayer, 2001).

3.4 Personal factors

The personal character of the farmer is important in the experimentation process. If he thinks he knows everything about his/her exploitation and that agricultural production can not be improved in any way, he will not experiment with alternatives; i.e., the farmer has prejudices that prevent him from experimenting (Sumberg and Okali, 1997).

The experimenting farmers who successfully develop innovations or inventions usually have a strong personality (Reij and Waters-Bayer, 2001); they are capable of observing changes and analyzing and interpreting the results (Zigta and Waters-Bayer, 2001); they should also be capable of managing a high degree of uncertainties (Rogers, 1995). Creativity, perseverance and conviction that the purpose will be successful, are qualities that favor the experimentation with alternative methods and technologies (Zigta and Waters-Bayer, 2001). In addition, innovators are usually curious, proud and willing to take risks (Critchley, 2000).

4. Sources of farmers' experiments

The experience acquired by a farmer throughout his/her life widens his/her local knowledge and is a good base for experimenting with new ideas or technologies (Zigta and Waters-Bayer, 2001). Successful experiments emerge by the combination of new ideas with local knowledge; the former can be introduced by extension workers, research centers or other farmers, but they can also be the experimenting farmer's own ideas (Bunch, 1991; Bentley, 2006).

Sumberg and Okali (1997) identified three important sources for farmers' experiments: in the first place, the farmer attempts something he/she observed or was recommended by others; in second place are the farmer's own ideas; and in the third place he/she experiments with technologies or methods that were actively promoted by institutions.

5. Motives of the experimenting farmers

Experimenting farmers can be motivated by economic considerations, for example, market demand. Through the experiments they try to increase incomes (Bentley, 2006) or avoid economic losses (Quiroz, 1999; Critchley, 2000). The reduction of the use of synthetic pesticides, and consequently of their cost, can lead to experiments with methods of biological control of pests. To decrease the cost and work time encourages farmers to experiment with new technologies or methods (Bentley, 2006). The need of a certain crop for self-consumption can promote experiments with new plants; so that there is no need to buy it in the market (Quiroz, 1999).

In addition to economic stimuli, personal stimuli can be identified for farmers' experimentation. The concern for the development of the exploitation, by later generations and the community, is another motivation for experimenting (Zigta and Waters-Bayer, 2001). There are experimenting farmers who look for a challenge and try something different, to be able to convince their neighbors afterwards (Scheuermeier, 1997). Through experimentation they discover a fundamental function of agriculture: to protect, create and improve land (Hocdé, 1997).

6. Characteristics of farmers' experiments

Farmers' experiments have general characteristics in common, although it is important to acknowledge that they depend on several factors and are different in each region (Quiroz, 1999).

During the process of agricultural production farmers go through several stages, in which they must make decisions and undertake actions to reach their objectives, as well as reflect about the results in order to improve them (Sumberg and Okali, 1997; Stolzenbach, 1999). In each stage of production in which the farmer must make decisions, a possibility for experimenting can appear. The exploitation management is a series of experimentations by means of which agricultural production should improve; it means that experimentation is an integral and continuous element of agriculture (Stolzenbach, 1999).

The experiments that require thorough changes in the organization of the agricultural system or the social relationships of the farmer decrease the willingness to experiment, which can be due to the complexity of the purpose or the risk that accompanies the experiment (Sumberg and Okali, 1997; Padel, 2001).

Such experiments are usually based on trial-error (Bajwa, Gill and Malhotra, 1997; Rajasekaran, 1999; Bentley, 2006); it means that farmers decide, according to the procedure, how to continue the experiment and can modify the methods. If the result fulfils the farmers' needs and improves their work or living conditions, it can be interesting for others too (Bajwa et al., 1997). The experiments are carried out with the available physical and biological resources, such as, for example, local seeds, manure, land or labor (Rajasekaran, 1999).

Farmers' experiments vary from very easy to very complex (Hocdé, 1997), but for decreasing the risk of experimentation the farmer usually applies new methods to small plots and maintains the experiment simple (Connell, 1991). There are few examples in which farmers' experiments cause radical and complex changes in the production system (Sumberg and Okali, 1997).

The farmer begins with an experiment to start a change in his/her agricultural system, but he/she generally does not have a concrete concept of the result; according to what is obtained he/she decides whether this type of experiment continues (Stolzenbach, 1997). Innovations, born from successful experiments, can cause changes in the work methods or the agricultural system. If they are complex, but successful at the same time, they can be quickly disseminated (Niemeijer, 1999).

6.1 Planning of farmers' experiments

Sumberg and Okali (1997) classify experiments as proactive and reactive. Most experiments are usually proactive, i.e., the farmer uses a certain statement before experimenting. Being proactive the farmers can experiment actively to solve the problems, trying several choices. Reactive experimentation is based on chance; it means that the farmer experiments without having a hypothesis or a statement.

Farmers do not usually analyze in detail their agricultural situation to formulate the justification, hypothesis and methods of their experiments afterwards; this process is developed intuitively. They can just have the curiosity to attempt something new, or have identified a problem and look for a solution; it can be also that they heard new information in which they see a unique opportunity to improve their situation (Scheuermeier, 1997).

6.2 Classification of farmers' experiments

The experiments can be classified according to their origin, cause or the topic selected by farmers to experiment. They can be also distinguished by the process followed in experimentation and the final result to which they arrive.

There is a combination of the above-mentioned types: double experimentation. It refers to cases in which farmers experiment with more than one variable at the same time; for example, manure varieties with different planting distances (Quiroz, 1999).

6.2.1 Classification according to the source of experiments

The sources of the experiments can be: the interest for solving the problems, curiosity or testing of expectations.

Farmers look actively for solutions for new or old problems of their agricultural system, motivated by the need or the wish to solve them, because they feel compelled to improve their current situation (Rhoades and Bebbington, 1991; Hocdé, 1997; Sumberg and Okali, 1997; Quiroz, 1999; Zigta and Waters-Bayer, 2001). Frequently, to acknowledge a problem or a need is the first step for the development of an innovation (Rogers, 1995).

As all human beings, farmers are curious and want to know if their ideas work. This type of "experiment due to curiosity" is carried out, for example, when a farmer obtains seeds from his/her neighbor and tries to sow them in his/her exploitation system (Stolzenbach, 1997; Quiroz, 1999; Zigta and Waters-Bayer, 2001).

The farmer can have an expectation of the results before beginning to experiment; the intention is to test a hypothesis, which he/she formulated cognitively. He/she does not have to be conscious of being testing a hypothesis, but he/she does it intuitively (Stolzenbach, 1997; Zigta and Waters-Bayer, 2001; Bentley, 2006).

6.2.2 Classification according to the topics of the experiments

There are topics that seem common and interesting for experimenting farmers. The experiments can be classified into: technical, economic, social and institutional, although 75% of those described in literature are technical (table 1). Within this category the experiments with new crops or varieties, soil preparation and fertility, sowing methods and crop density are the most common topics (Sumberg and Okali, 1997; Nielsen, 2001). Complex experiments, such as the social or institutional ones, require high management and organization capacities, for which they are scarce (Sumberg and Okali, 1997).

6.2.3 Classification according to the experimentation process

According to several authors, adaptation experiments and the ones originated from other experiments can be distinguished.

Adaptation experiments often occur after the introduction of an innovation. In this sense, the experimentation process arises from adopting and adapting innovations. The modification or reinvention is important for farmers (Sumberg and Okali, 1997) because almost all of them experiment with adopted technologies (Cramb, 2005). Adapting innovations means that farmers apply new elements and factors to their agricultural system; adaptation is a complex process of experimental learning (Mak, 2001; GebreMichael, 2001). Adaptation experiments can be found in two forms (Rhoades and Bebbington, 1991): a) farmers that apply and modify an innovation in a known environment; b) farmers that apply a known technology in a new environment. Adaptation is defined as the degree in which an innovation is changed by its user during the adoption and introduction process (Rogers, 1995).

On the other hand, farmers' experiments do not have to appear isolated, but can cause a series of experiments or innovations closely and logically connected to each other, called in literature `experiments that arise from other experiments'. The introduction of a new technology or an alternative method can inspire the farmer to experiment more and thus adjust the agricultural system to the changes caused by such method (Tchawa, 2001). When the conditions of an agricultural system change, as consequence of an experiment or an innovation, the farmers must adjust to the new situation, and thus an experiment or innovation can trigger other experiments. For example, an innovation that increases agricultural production requires faster harvest methods and a better distribution system (Reij and Waters-Bayer, 2005).

6.2.4 Classification according to the final result of the experiment

From farmers' experiments `hard innovations' can emerge, which are physical and visible results, e.g. new tools, different substances for pest control, soil fertility, crop rotation, sowing technology, animal rearing, irrigation, drainage, etc. (Rogers, 1995; Waters-Bayer, 2005).

In addition, `soft innovations' can emerge, which mean that the result of the experiment is a method for improving an intangible situation. They can be: knowledge, ability, procedures and/or principles that are useful as informative base for the development of tools or technologies. Soft innovations appear when farmers experiment with new marketing or communication methods (Rogers, 1995; Waters-Bayer, 2005). They can also be qualitative methods for pest counting in agricultural plots, important for effective and sustainable control (Bentley, 2006).

7. Evaluation of the experiments

Evaluation, as well as reflection about the results by experimenting farmers, is supported in a wide context of agricultural production. The agricultural system is the base for the family livelihood, which makes farmers perceive it as a highly valuable system, although it does not mean that they attempt to fulfill their expectations. Besides, the environmental surroundings prevents total manipulations and the farmers would not have advantage if they changed variables for improving the results (Sumption, 2004); they are usually self-critical, and the fact that they work in the agricultural system allows them to observe continuously the experimentation process (Stolzenbach, 1999; Sumption, 2004).

In many cases the experiments are qualitative, i.e. without sample groups or numbers (Bentley, 2006), although other sources indicate that up to 40 % of the experimenting farmers use control groups or direct comparison to test their ideas. Some trust their local knowledge as the "historical control", i.e. farmers that have a deep knowledge, arisen from experience and also know which factors have influence (Sumberg and Okali, 1997).

8. Differences between farmers' experiments and scientific experiments

Farmers experiment and innovate with their own methods, which are normally different from the scientific ones, and do it under different conditions from researchers (Sumberg et al., 2003). The farmer is part of the system in which he/she is experimenting and has a direct interest in improving the situation regarding his/her needs. Although farmers sometimes change variables during the experimentation process, they assure they can determine the limiting factor. Scientists usually reduce reality to few variables and, of course, they are more thorough. The methods used by scientists must be understandable, which allows them to explain their results to a group of experts. According to Stolzenbach (1999), the model of scientific experimentation is too strict for farmers.

9. Farmers' experiments, innovation and ecological agriculture

Until the last decade of the past century, the first ecological farmers in Europe did not have the support of science or consultancy or agricultural extension programs. They had to develop ecological agriculture individually, through experiments and continuous innovations. Institutional research has denied for many years the efforts of ecological farmers (Padel, 2001).

Farmers preferred to experiment with ecological methods before making the conversion, to test little by little the feasibility of ecological agriculture and reduce the risk. The experiments in small plots, as for example family gardens, before the conversion to ecological agriculture, reduce the technical and economic risk. Such experiments can include: the reduction of fertilizers, use of alternative treatments for animals, introduction of legumes, new crops under ecological production or conversion of only a few plots (Padel, 2001; König, 2003).

Based on the results of the experiments, farmers made decisions for changing their way of working. Most of them have some experiences with ecological methods before the conversion to ecological agriculture. But not only before this process does experimentation have an important role, but also during the first years (Padel, 2005). The conversion requires deep and complex changes in the agricultural system, which are usually accompanied by an intense learning process to obtain experiences in ecological production (Sumption, 2004).

Conclusions

Farmers' experiments increase self-consciousness, strengthen their identity and confidence in their capacities (Hagmann et al., 1997), because they perceive they are capable of experimenting and developing useful technologies or methods. Experimentation is a basic process for the development of agriculture, through which farmers have the opportunity to increase their experience and widen their local knowledge (Sumberg and Okali, 1997).

The integration of such experiments in the agenda of scientific research would improve the cooperation between scientists and farmers, and the results would be more adequate for the agricultural community (Sumberg and Okali, 1997).

Experimenting farmers can be conscious or not of their research and learning activities (Hocdé, 1997); every time they begin using an unknown crop or method, they enter a new learning process (Sumption, 2004). Through the experiments farmers understand better the agricultural system, understand why they try to change certain variables and at the same time reflect on experimentation and its results (Hocdé, 1997; Stolzenbach, 1999).

Experiments are actively used as a learning tool for the farmers to acquire knowledge of their surroundings and production system; through this learning method they increase their consciousness and pride.

In addition, experimentation can lead to social acknowledgement with a generalization of its use in the community. Farmers develop a new willingness to experiment and the quantity of experiments increases (Hagmann and Chumab, 2002). Through the experiments they obtain a deep understanding of their surroundings, which is continuously changing (Niemeijer, 1999).

A farmer's innovation, born from a successful experiment, can activate the willingness to experiment within the community. In addition, there are innovations that require the help of other farmers and thus a social network of experimenting farmers is enhanced (Tchawa, 2001).

For all these reasons, farmers' experimentation is an important and indispensable practice for successful rural development, and also for the dissemination of the results of scientific agronomic studies. In general, each farmer incorporates new recommendations that emerge from scientific research, through experiments in his/her farm. For such reason scientific experimentation does not end after achieving and publishing the results.

Experimentation in agronomy is a continuous process carried out in experimental plots and farms by scientists, and it is also performed in farms by the farmers themselves. From the farmers' experiments often emerge new ideas, which are incorporated to scientific research.