resistance to abiotic pressures may allow crops to be grown in marginal, low productivity areas, while increased storage stability and delayed ripening will benefit those with few resources to invest in refrigeration and other equipment to increase the shelf life of agricultural produce. Micronutri-ent-enriched staple crops have also been developed to target the most vulnerable—resource-poor women and children (Combs Jr. et al., 1996; Bouis, 2000).
     There is increasing interest in the potential for transgenic plants to produce pharmaceuticals and vaccines through molecular farming (pharming). Vaccines can be used to pre­vent or combat many of the diseases that cause illness and death in low-income countries, but are expensive, must be refrigerated and administered by trained personnel and re­quire sterile delivery through needles that are often difficult and expensive to obtain. Although research in this area is in the early stages, vaccines against some infectious gastro­intestinal diseases have been produced in potatoes, bananas and corn (Thanavala et al., 1995; Lamphear et al., 2004). Transgenic plants are also being evaluated for a variety of non-food applications, including bioremediation, modifica­tion of fiber content and biodegradable plastics (Haigler and Holaday, 2001; He et al., 2001; Poirier, 2001; Scheller et al., 2001).
     Genetic use restriction technologies (GURT) are based on regulating gene expression to restrict plant propagation from a second generation of seed. Unlike variety-specific (V-GURT; "terminator technology") which results in sterile seed, specific trait, or T-GURT would enable farmers to save their own seed, but lack access to the added traits in the absence of payment for chemical activators. In addition to their use restriction properties discussed above, GURTs rep­resent an environmental containment strategy through their ability to eliminate the spread of transgenic seed (V-GURT) or transgenes to wild plant relatives (T-GURT).
     Potential productivity advantages from GURTs include T-GURTs enabling a farmer to restrict the expression of a trait when there is an advantage in doing so; for example, at a specific phase of development or during periods of bi-otic or abiotic stress (FAO, 2001). Productivity gains from V-GURTs include the ability to safeguard the integrity of adapted maternal breeds or to prevent preharvest sprouting. As with any growth regulator applied to crops, environmen­tal or human health issues may be associated with the use of chemical activators (i.e., tetracycline, copper, steroids) and hence the effects of these need to be addressed.
     Transgenic crops and GURT offer a promising means to increase agricultural productivity in cropping systems. However, these technologies have the potential to affect human and animal health, have substantial social and eco­nomic impacts on grower communities and can significantly alter agricultural ecosystems through effects on the environ­ment. Despite human health concerns, several studies with animal models and a range of transgenic crops have failed to show any toxic, allergenic, or nutritional effects of the transgenic crop tested (Noteborn et al., 1995; Hammond et al., 1996; Harrison et al., 1996; Hashimoto et al., 1999ab; Folmer et al., 2000; Momma et al., 2000; Sidhu et al., 2000; Teshima et al., 2000; Ash, 2003; Donkin et al., 2003; Stan­ford et al., 2003; Hammond et al., 2004; Kan and Hartnell, 2004). These findings in no way negate the need to apply

rigorous standards to health risk assessments of individual technologies; in addition, comparative risk assessments with alternative control regimes will be needed.
     Transgenic and genetic use technologies have the poten­tial to increase economic returns via improved crop yields under stress conditions and reduced chemical inputs, while preventing the spread of transgenes in the case of GURT applications. Transgenic technology can significantly affect the cultural and economic situations of producers—as can conventional plant breeding—but at a faster rate than the latter. The threats of biodiversity reduction through "ge­netic pollution" and "superweed" creation are scenarios with  far-reaching  negative  consequences  for  livelihoods and cropping systems. Further, the technologies are expen­sive and commit farmers to regular financial outlays for improved seed or chemicals each season that may not be achievable.
     Potential environmental effects of transgenic technol­ogy include: (1) adverse effects on non-target organisms, (2) gene flow into wild plant communities or soil organisms and (3) development of resistance by target pests. Non-target entomophagous insects and parasitoids are invaluable in in­tegrated pest management approaches employed to control outbreaks of insect pests. Most of the insecticides used for the control of rice stem borers and leaffolders have been found to harm beneficial insects, while multiple Bt rice lines show no significant non-target effects (Chen et al., 2006). An evaluation of direct toxicity or indirect food chain-re­lated effects on a large variety of insects and crops indicates no adverse impacts (Lopez and Ferro, 1995; Orr and Lan-dis, 1997; Pilcher et al., 1997; Riddick and Barbosa, 1998; Volkmar et al., 2000; Zwahlen et al., 2000; Bourguet et al., 2002; Cowgill et al., 2002; Al-Deeb et al., 2003; Cowgill and Atkinson, 2003; Dutton et al., 2003; Jorg et al., 2003; Volkmar et al., 2003). Although most of the evidence sug­gests that transgenic crops do not have direct effects on beneficial insects, adverse effects of Bt proteins on benefi­cial insects via compromised food quality of their prey have been reported (Hilbeck et al., 1998; Schuler et al., 1999; Meier and Hilbeck, 2001; Ponsard-Sergine et al., 2002) and transgenic corn resulting from event 176 had adverse effects on Lepidoptera (butterflies), arguing for case-by-case evalu­ations (Losey et al., 1999; Jesse and Obrycki, 2000; Hell-mich et al., 2001; and Stanley-Horn et al., 2001; Zangerl etal.,2001).
     Although a variety of transgenic crops have demon­strated changes in microbial, protozoan and nematode pop­ulations in soil (Donegan et al., 1995; Di Giovanni et al., 1999; Donegan et al., 1999; Griffiths et al., 2000; Lukow et al., 2000; Marroquin et al., 2000; Cowgill et al., 2002), the data are difficult to interpret and tie to ecosystem function and a large number of studies have shown no effect on these soil populations or their processes (Al-Deeb et al., 2003; Blackwood and Buyer, 2004; Devare et al., 2004; Wu et al., 2004; Fang et al., 2005; Devare et al., 2007).
     Gene flow between crops and their wild relatives is com­mon and, between plants capable of hybridizing, inevitable if grown within the crop's pollen dispersal range (Ellstrand et al., 1999). Pollen-mediated crop-to-crop transgene flow in rice can be maintained at negligible levels with short spa­tial isolation (Rong et al., 2007). Insect resistance to Bt has