mine market certifications and reduce farmer profits [Global Chapter 6]. Seed supplies and centers of origin may be put at risk when they become mixed with unapproved or regulated articles in source countries [Global Chapter 3].

Trees and crops
Plant breeding and other biotechnologies (excluding trans-genics discussed below) have made substantial historical contributions to yield [Global Chapter 3]. While yield may have "topped out" under ideal conditions [Global Chapter 3], in developing countries the limiting factor has been ac­cess to modern varieties and inputs instead of an exhaus­tion of crop trait diversity [Global Chapter 3], and therefore plant breeding remains a fundamental biotechnology for contributing to sustainability and development goals.
     Biotic and abiotic stresses, e.g., plant pathogens, drought and salinity, pose significant challenges to yield. These chal­lenges are expected to increase with the effects of urbaniza­tion, the conversion of more marginal lands to agricultural use [SSA Chapter 1], and climate change [CWANA Chapter 1; Global Chapter 7; SSA Chapter 1]. Adapting new culti-vars to these conditions is difficult and slow, but it is again plant breeding perhaps complemented with MAS, that is expected to make the most substantial contribution [Global Chapters 3, 6]. Genetic engineering also could be used to introduce these traits [Global Chapter 5; NAE Chapter 6]. It may be a way to broaden the nutritional value of some crops [ESAP Chapter 5]. If GM crops were to increase pro­ductivity and prevent the conversion of land to agricultural use, they could have a significant impact on conservation [Global Chapter 5]. However, the use of some traits may threaten biodiversity and agrobiodiversity by limiting farm­ers' options to a few select varieties [ESAP Chapter 5; Global Chapters 3, 5, 6].
     Breeding capacity is therefore of great importance to assessments of biotechnology in relation to sustainability and development goals [NAE Chapters 4, 6]. In develop­ing countries, public plant breeding institutions are common but IP and globalization threaten them [Global Chapters 2, 6]. As privatization fuels a transfer of knowledge away from the commons, there is a contraction both in crop diversity and numbers of local breeding specialists. In many parts of the world women play this role, and thus a risk exists that privatization may lead to women losing economic resources and social standing as their plant breeding knowledge is ap­propriated. At the same time, entire communities run the risk of losing control of their food security [CWANA Chap­ter 1; Global Chapter 2].
     Plant   breeding   activities   differ   between   countries, so public investment in genetic improvement needs to be augmented by research units composed of local farming communities [Global Chapters 2, 6]. In addition, conflicts in priorities, that could endanger in situ conservation as a resource for breeding, arising from differences in IP protec­tion philosophies need to be identified and resolved [Global Chapter 2]. For example, patent protection and forms of plant variety protection place a greater value on the role of breeders than that of local communities that maintain gene pools through in situ conservation [Global Chapter 2]. It will be important to find a new balance between exclusive access secured through IPR or other instruments and the

need for local farmers and researchers to develop locally adapted varieties. It will be important to maintain a situa­tion where innovation incentives achieved through IPR in­struments and the need for local farmers and researchers to develop locally adapted varieties are mutually supportive. Patent systems, breeders' exemptions and farmers' privilege provisions may need further consideration here  [Global Chapter 2]. An important early step may be to create effec­tive local support for farmers. Support could come from, for example, farmer NGOs, where appropriate, to help develop local capacities, and advisers to farmer NGO's to guide their investments in local plant improvement. Participatory plant breeding, which incorporates TK, is a flexible strategy for generating new cultivars using different local varieties. It has the added advantage of empowering the local farmer and women [Global Chapter 2]. A number of ad hoc private ini­tiatives for donating or co-developing IP are also appearing [Global Chapter 2], and more should be encouraged.
     The decline in numbers of specialists in plant breeding, especially from the public sector, is a worrisome trend for maintaining and increasing global capacity for crop im­provement [Global Chapter 6]. In addition, breeding supple­mented with the use of MAS can speed up crop development, especially for simple traits [Global Chapter 3; NAE Chapter 6]. It may or may not also significantly accelerate the de­velopment of traits that depend on multiple genes [Global Chapter 6]. Provided that steps are taken to maintain lo­cal ownership and control of crop varieties, and to increase capacity in plant breeding, adaptive selection and breeding remain viable options for meeting development and sustain­ability goals [Global Chapter 6; NAE Chapter 6].

Gene flow
Regardless of how new varieties of crop plants are cre­ated, care needs to be taken when they are released because through gene flow they can become invasive or problem weeds, or the genes behind their desired agronomic traits may introgress into wild plants threatening local biodiver­sity [Global Chapter 5]. Gene flow may assist wild relatives and other crops to become more tolerant to a range of en­vironmental conditions and thus further threaten sustain­able production [Global Chapters 3, 6]. It is important to recognize that both biodiversity and crop diversity are im­portant for sustainable agriculture. Gene flow is particularly relevant to transgenes both because they have tended thus far to be single genes or a few tightly linked genes in ge­nomes, which means that they can be transmitted like any other simple trait through breeding (unlike some quantita­tive traits that require combinations of chromosomes to be inherited simultaneously), and because in the future some of the traits of most relevance to meeting development and sustainability goals are based on genes that adapt plants to new environments (e.g., drought and salt tolerance) [Global Chapter 5].
     Transgene flow also creates potential liabilities [Global Chapter 6]. The liability is borne when the flow results in traditional, economic or environmental damage. For exam­ple, the flow of transgenes from pharmaceutical GM food crops to other food crops due to segregation failures could introduce both traditional and environmental damage. An important type of potential economic damage arises from