lated from any possibility of cross-fertilization or genetic contamination by uniform GE crops, is necessary as genetic uniformity or changes in the genetic integrity of local vari­eties could have considerable impacts. Moreover, biologi­cal and cultural diversity and the associated local skills and resources, are needed for rural populations to maintain or recover production processes.
     Furthermore, the maintenance of traditional agroeco-systems is a sensible strategy to preserve in situ repositories of crop germplasm (Altieri, 2003). However, this cannot be done in isolation from the maintenance of sociocultural or­ganization, including of the need to organize small farmers into groups to strengthen their collective bargaining posi­tions, particularly in facing corporate players (see 3.4.5). The process must be linked to rural development efforts that give equal importance to local resource conservation, food self-sufficiency and some level of market participation. In order for peasants to have a competitive edge, they need to be able to produce "unique" agricultural crops (i.e., GE-free) for niche markets. Such "uniqueness" is crucial for maintaining the stability of local farming systems in times of uncertainty.
     AKST for sustainable agriculture should thus fully in­volve farmers and develop technologies that are low-cost, readily available and responsive to diverse local conditions, without posing risks, particularly to the diversity base of poor farmers. It is difficult to see how traded GE commodity crops can meet these criteria.

3.4.6     Fisheries and aquaculture
The liberalization of trade has led to a big increase in ex­ports of fish and fish products from developing countries, as a whole and Asia in particular. Fisheries now generate more foreign exchange than any other traded food commodity, such as rice, coffee, tea or cocoa (FAO, 2004b) (see 2.2.4).
     However, there are changes within this trade, though the geographical pattern remains the same. First, there is the shift from export of raw material to be processed in developed countries to export of processed fish. The devel­opment of fish processing capacity and knowledge in devel­oping countries of Asia has enabled them to bring about a shift in the location of processing. The lower wages in Asian countries compared to the former processing countries (EU, Japan and US) has facilitated this shift in location. More­over, the highly perishable nature of fish also favors the shift of processing to the source of raw materials. There is also a learning or capability-building process, whereby labor and management in Asia have learnt and invested in the technol­ogy of processing.
     Second, there is also a shift to exports of live fish. Most of it is for ethnic markets in the developed countries. The migration of large numbers of Asians has led to the growth of a market for live fish from their countries of origin. Some of the live fish is also of the ornamental variety for aquari­ums. In both cases, the development of transport and logis­tics technology have enabled a growth in this sector of trade, which now accounts for about 10% of fish trade.
     With the growing world fish trade and the possibility of reasonably elastic export earnings, there were initial trends towards over-exploitation offish resources. At least 25% of fish varieties in the world are reported to be substantially

over-exploited. There was an increase in the proportion of overexploited and depleted stocks from around 10% in mid-1970s to about 25% in early 2000s (FAO, 2004b). Be­sides bans on fishing, often brought about by the collapse of certain sectors, such as cod in the North Atlantic, there have also been technological shifts towards aquaculture, both of the freshwater and marine varieties. This is a very major technology change in response to the growing demand for fish along with the relatively fixed fish resource available (see 2.2.4).
     While the conduct of aquaculture has its own problems, which will be dealt with later, it has certainly enabled a growth of production without endangering available stocks of wild fish, as trade based on capture fisheries tends to do. So far, in Asia, aquaculture has developed substantially for freshwater fish. Hatchery-based marine aquaculture is not as developed. Most marine aquaculture, as for prawn and seaweed, still depends on collection of seed from the wild. The jump to true aquaculture, with hatchery rearing of fry, has yet to be developed for many marine species. The type of marine aquaculture developed for salmon and trout has yet to be developed for the fishes of Asia. Recently, Japan has developed technology for sustaining bluefin tuna broodstock in offshore cages, leading to the first closed-cycle breeding of tuna. In Indonesia, fishers are replacing cyanide harvesting of reef fish with hatchery-raised juveniles of aquarium fish.
     As pointed out earlier, most of the global fish trade is from developing to developed economies. There are some technology and production concerns arising from this spe­cific nature of trade. There are also other concerns arising from other aspects of global trade.
     Two concerns that arise from the developing to devel­oped nature of the trade relate to the meeting of quality standards, particularly those of Sanitary and Phytosanitary Standards (SPS) (see 3.2). A not entirely unrelated matter is that of traceability, something insisted on by the developed country fish retail chains that have to contend with supplier responsibilities.
     SPS problems have led to many temporary bans on imports of fish from Asian countries, particularly shrimp from various countries. Fish and fish products represented the largest category, above 25%, of food safety and quality alerts in the EU. Frequently there have been bans on imports of fish from various Asian countries. Initially they were met with cries of trade barriers. But after some time, the various Asian countries have begun to take measures to comply with these SPS standards. While they add to cost, the reduction of pesticide or veterinary drug residues or elimination of growth hormones are certainly desirable in themselves.
     With a large part of Asian aquaculture being carried out in small farms, traceability is certainly a problem. But as an example from Bangladesh, the Noakhali Gold Proj­ect, shows, this can be tackled along with that of meeting SPS standards by linking groups of small producers with the larger processing and packaging units. The meeting of SPS and traceability, however, is more a matter of management methods than one of technology. This intervention was pro­moted by a donor-funded project.
     Consumers in many parts of the world are concerned about the ecological impacts of different types of fishing and aquaculture. Endangered and charismatic species, like the