situation, one option would be to introduce an institutional model that gives these initiatives greater visibility, connectivity and sustainability, while providing stable funding and monitoring the quality of the proposed activities and the competence of participants. An institution that is standardized at the national or regional level, based on the model of the primary or secondary school, with branches in all municipalities, can accomplish these goals.

4.3.2 Adoption of a participatory approach to research
Mainstreaming the participatory approach to systems (styles) of research requires the participation of different actors in the research activities. This is so in order to incorporate local knowledge in combination with scientific knowledge and to seek alternative and common solutions to the problems of producers. This will ensure that producers resolve their own problems, master new technologies and increase their knowledge and awareness of the problems currently being faced (Schmitz and Mota, 2006). To this end, different participatory methodologies and tools (GTZ, CIAT, ASPTA, among others) that have produced good results in different countries should be used. The use of participatory methodologies in research and extension projects has proven to be of great use both for the potential to recover local knowhow and for the acquisition of new knowledge, generated in conjunction with scientists. The adoption of technology can then be done much faster and more efficiently (EMBRAPA, 2006; Pérez, et al. 2001).
     This means that, for example, greater use will be made of participatory approaches in such areas as the selection of plant varieties and in field tests for new crops that have potential for diversification. Achieving this goal will require significant restructuring of the national research institutions in many countries of Latin America and the Caribbean to put greater emphasis on disbursement mechanisms for research funding and on the training of personnel in participatory methods. The response capacity of research systems must be expanded to meet the demands of the market and while hybrid genetic material might be acceptable for use in diversification initiatives, material that can be replicated on farms is likely to be required for traditional crops (Dixon et al., 2001).

4.3.3 Strengthening of R&D networks
The creation of specialized networks in certain sectors or cultivation of specific crops may be another option for strengthening interaction between countries of the region and between knowledge systems. The creation of networks requires training and the generation of knowledge by the direct participants, namely local producers and consumers, and the establishment of mechanisms for the organization (generated by AKST systems) of small- and medium-sized producers.

     There currently already exists within the international scientific community and among donors a recognition that both organized actors and research centers should develop projects that are more directly geared to the generation of technologies and products that contribute to the reduction of poverty, with priorities being subject to change (CGIAR, 2003). Networks must also be established to protect and

disseminate innovations that benefit rural populations in accordance with their local conditions and which help to institutionalize knowledge sharing arrangements (Durston, 2002).
     Regional and above all subregional cooperation, where planting, soil and climate conditions are more roughly similar, should be significantly strengthened. Knowledge sharing should also be encouraged between other actors in subregional and national innovation systems, in addition to NARIs such as universities, NGOs, cooperatives and producer and private sector associations (Bisang et al., 2000).
     A recent report on agricultural research and development cooperation programs in Latin America and the Caribbean clearly points to a lack of coordination between regional initiatives and the need for the organization of similar networks and governance structures for research and development and innovation activities. Cooperative programs, such as Procis (PROCITROPICO, PROCISUR, PROCIANDINO, PROCICARIBE; see option 15 under Key Issues) are increasingly concerned with the organization of research networks and partnerships (FORAGRO, 2006; Salles-Filho, 2006).
     The evaluation of these programs implemented at various levels (Evenson y Cruz, 1989; Cruz y Avila, 1992; Avila et al., 2005; Salles-Filho et al., 2006a,b,c) shows that cooperation has resulted in a great deal of spill-over between countries. In other words, neighbouring countries benefit from research undertaken on the other side of the frontier. Evaluations have also shown that these programs need to be evaluated, restructured and extended to other actors so that they could more effectively fulfil their other objectives (Salles et al., 2006abc).

4.3.4 Organizational models
Currently the main challenge in nearly all Latin American countries is to build and strenghten their institutional capacities in order to promote the development of their AKST. While many countries of the region have made significant efforts to modernize the State, in terms of the first and second generation of reforms,23 the results were incomplete— particularly those of the second generation—and were not part of a coherent set of policies that could help to develop the capacity to create the minimum conditions necessary for the development of AKST in the region. Actors in the system, especially those in the public sector, on the whole suffer more from the absence or unpredictability of the flow of financial resources, the centralization with limited autonomy of centers/stations, deficiencies, low wages and rotation of qualified personnel and the lack of administrative and financial flexibility (Bisang et al., 2000).
     These problems are closely linked to the organizational
____________________
23 Initially in the so-called first generation of reforms the emphasis was placed on the objective of deregulation and waste reduction, the size of the state and its intervention in the economy. These reforms were carried out in the late 1980s and early 1990s (as one of the main pillars of the so-called Washington consensus). Later, in the second generation of reforms, emphasis was placed on building the capacity of the state.