for up to one-third of the patentable intellectual property developed by the department, with the University retaining the patent rights and earning royalties from the patents. Participating faculty, in turn, received access to proprietary databases held by Novartis (Berdahl, 2000; Busch et al., 2004). The Novartis agreement disquieted those who believed it indicated a transition toward the privatization of public universities; critics argued that by allowing Novartis to participate even as a minority vote on the funding committee, the University was allowing a private company to chart the course of research at the University (Berdahl, 2000). Others pointed out that faculty members applying for research support from the federal government possibly also tailored their applications to increase their chances of support. This situation illustrates the need for codes of conduct in universities to guide their interactions with industry (Washburn, 2005) in order to preserve independence and capacity to deliver disinterested public goods and maintain public trust (Vilella et al., 2002). More public-private partnerships without ensuring such codes may reduce the space for public interest science (Washburn, 2005), although under certain conditions, university partnerships with private actors may contribute to equitable and sustainable development. For instance, the Seed Nursery at the Faculty of Agronomy, Buenos Aires University (www.agro.uba.ar) and the Argentine Agrarian Federation (www.faa.com.ar) developed high-yielding non-Bt corn hybrids (FAUBA 207, 209 and 3760), which are locally adapted and affordable by small-scale farmers and were released to market at less than half the price of the main competitors (Vilella et al., 2003; Federacion Agraria Argentina, 2005; http://www.todoagro .com.ar/todoagro2/nota.asp?id=6542).

(3) Agrotechnical institutes. Postsecondary institutes that are not part of the university system usually depend on public funding from Ministries of Education or Agriculture. They mostly train technicians in agricultural competences related to local labor demand in order to bridge the gap between untrained farmers, semi-skilled technicians and university graduates. However, many developing countries have given little attention to the training demanded by agricultural service agencies and agroindustries. Other countries, such as India or Brazil, invested heavily in such training. In Brazil, the Federal Centers of Technological Education (CEFETs) originated in agrotechnical or technical schools that were upgraded to tertiary-level institutes in the mid-1990s. They have developed good links with the private sector and sometimes share resource training activities through "sandwich courses." They have become drivers for the application of technology, but an extension worker with certificate-level training and field experience can seldom bridge to a degree program (Atchoarena and Gasperini, 2002; Plencovich and Costantini, 2003). The Sasakawa Africa Fund for Extension Education (SAFE) specifically addresses this need in SSA. Other countries have chosen an alternative agricultural school system shaped after the Maisons Familiale Rurale (rural family house) (Granereau, 1969; Forni et al., 1998; García-Marirrodriga and Puig Calvó, 2007). Today there are more than 1,300 such schools in forty countries, alternating residential training and experience on the family farm. In Argentina, a large group of secondary public

schools managed privately by NGOs, foundations, and other private actors and federated under the umbrella of an apex organization (FEDIAP; http://www.fediap.com.ar/) manages 3,000 teachers and about 15,000 students, taking occupational education deep into marginal and vulnerable areas (Plencovich and Costantini, 2006).

2.2.3 Producers of AKST at regional and international levels

The institutional arrangements for development-oriented AKST at international levels have evolved from rather simple relationships organized largely by and in support of colonial interests, through focused support organized through the CGIAR system largely under the guidance of multilateral and bilateral development organizations, to arrangements that are rapidly diversifying under market pressures. The increasing attention to environmental issues, especially from the early 1980s onwards, also gave rise to arrangements that made effective use of collective capacity to address shared practical and policy problems related to such issues as watershed management, vector-borne diseases and biodiversity conservation efforts. Examples include CSOs, such as the South East Asian Regional Initiatives for Community Empowerment (SEARICE) in the Philippines (cf. http://www .searice.org.ph/), which serves as the secretariat for regionwide advocacy, lobbying and action among networks of CSOs to promote and protect farmers' seed exchanges and sales and to ensure legal recognition of farmer-bred varieties and of community registries of local plants, animals, birds, trees, and microorganisms. SEARICE has become a major actor in the establishment of community-based native seeds research centers, such as CONSERVE in the Philippines and farmer field schools for plant genetic resource conservation development and use in Laos, Bhutan, Vietnam and community biodiversity conservation efforts in Vietnam, Thailand, and the Philippines. SEARICE today is recognized as an effective and legitimate partner in sustainable and equitable development. The Mekong River Commission (MRC) offers a different kind of arrangement; founded in 1995 by the Agreement on the Cooperation for the Sustainable Development of the Mekong River Basin (http://www.mrcmekong .org/). It is funded by contributions from the downstream member countries (Cambodia, Laos, Thailand, Viet Nam) and donors and is considered an important institutional innovation that is successfully bringing together cross-sectoral knowledge and helping actors to learn from policy experiments. However economic drivers within the member states resulting in upstream development of irrigation and hydroelectric power in China are undermining local efforts to forge more sustainable development pathways (Jensen, 2000; MRC, 2007). In SSA regional AKST arrangements have emerged and today their NARIs also act as regional service centers. ASARECA and CORAF were established in the late 1990s in eastern and western Africa respectively to fill gaps and build on strengths but no assessment can yet be made of their effectiveness. In southern Africa the formalization of interstate collaboration in AKST has not yet occurred. The South African Agricultural Research Council and universities continue to provide a regional backup service and various R&D networks seek to fill some of the severe gaps in public and private capacity.