Based on the fact that more than 500 species of pests have already evolved resistance to conventional insecticides, pests can also evolve resistance to Bt toxins present in transgenic crops (Gold l994). No one questions if Bt resistance will develop, the question is now how fast it will develop. Susceptibility to Bt toxins can therefore be viewed as a natural resource that could be quickly depleted by inappropriate use of Bt crops (Mellon and Rissler l998) . However, cautiously restricted use of these crops should substantially delay the evolution of resistance. The question is whether cautious use of Bt crops is possible given commercial pressures that have resulted in a rapid roll-out of Bt crops reaching 8.2 million hectares worldwide in 2000.  The refuge strategy of setting aside 20-30 percent of the land to non-Bt crops to delay resistance is very difficult to implement regionally. Data from the Midwest shows that Bt corn saves on some insecticide use and  yields are 2.4 bu/acre higher than conventional corn but only under high European corn borer infestations (USDA 1999). On the other hand organic corn growers use no insecticides and obtain yields (4.8-9 t/ha) similar or slightly higher than conventional farmers (5.0-7.l t/ha) (National Research Council l984) .

Bacillus thuringiensis proteins are becoming ubiquitous, highly bioactive substances in agroecosystems present for many months. Most, if not all, non-target herbivores colonizing Bt crops in the field, although not lethally affected, ingest plant tissue containing Bt protein which they can pass on to their natural enemies in a more or less processed form. Polyphagous natural enemies that move between crop cultures are found to frequently encounter Bt containing non-target herbivorous prey in more that one crop during the entire season. This is a major ecological concern given previous studies that documented that Cry1 Ab adversely affected the predaceous lacewing Chrysoperla carnea reared on Bt corn-fed prey larvae (Hilbeck l998).

These findings are problematic for small farmers in developing countries who rely for insect pest control, on the rich complex of predators and parasites associated with their mixed cropping systems (Altieri 1994). Research results showing that natural enemies can be affected directly through inter-trophic level effects of the toxin present in Bt crops raises serious concerns about the potential disruption of natural pest control, as polyphagous predators that move within and between crop cultivars will encounter Bt-containing, non-target prey throughout the crop season. Disrupted biocontrol mechanisms will likely result in increased crop losses due to pests or to the increased use of pesticides by farmers with consequent health and environmental hazards.

The possibilities for soil biota to be exposed to transgenic products is very high. The little research conducted in this area has already demonstrated long term persistence of insecticidal products (Bt and proteinase inhibitors) in soil ( Donegan and Seidler 1999). The insecticidal toxin produced by Bacillus thuringiensis  subsp. kurskatki  remain active in the soil, where it binds rapidly and tightly to clays and humic acids. The bound toxin retains its insecticidal properties and is protected against microbial degradation by being bound to soil particles, persisting in various soils for at least 234 days (Palm et al l996). In another study researchers confirmed the presence of the toxin in exudates from Bt corn and verified that it was active in an insecticidal bioassay using larvae of the tobacco hornworm (Saxena et al l999). Given the persistence and the possible presence of exudates, there is potential for prolonged exposure of the microbial and invertebrate community to such toxins, and therefore studies should evaluate the effects of transgenic plants on both microbial and invertebrate communities and the ecological processes they mediate (Altieri 2000).

If transgenic crops substantially alter soil biota and affect processes such as soil organic matter decomposition and mineralization, this would be of serious concern to organic farmers and most poor farmers in the developing world who cannot purchase or don’t want to use expensive chemical fertilizers, and that rely instead on local residues, organic matter and especially soil organisms for soil fertility (i.e. key invertebrate, fungal or bacterial species) which can be affected by the soil bound toxin. Soil fertility could be dramatically reduced  if crop leachates inhibit the activity of the soil biota and slow down natural rates of decomposition and nutrient release. 

The available  independently generated scientific information suggests  that the massive use of transgenic crops poses substantial potential risks  from an ecological point of view. The ecological effects are not limited to pest resistance and creation of new weeds or virus strains (Kendall et al l997).. As argued her in, transgenic crops can produce environmental toxins that move through the food chain and also may end up in the soil and water affecting invertebrates and probably ecological processes such as nutrient cycling (Altieri 2000). No one can really predict the long term impacts that will result from such massive deployment of such crops.

Not enough research has been done to evaluate the environmental and health risks of transgenic crops, an unfortunate trend as most scientists feel that such knowledge was crucial to have before biotechnological innovations were upscaled to actual levels. There is a clear need to further assess the severity, magnitude and scope of risks associated with the massive field deployment of transgenic crops. Much of the evaluation of risks must move beyond comparing GMC fields and conventionally managed systems to include alternative cropping systems featuring crop diversity and low-external input approaches. This will allow real risk/benefit analysis of transgenic crops in relation to known and effective alternatives. 

Moreover, the large-scale landscape homogenization with transgenic crops will exacerbate the ecological problems already associated with monoculture agriculture. Unquestioned expansion of this technology in to developing countries may not be wise or desirable. There is strength in the agricultural diversity of many of these countries, and it should not be inhibited or reduced by extensive monoculture, especially when consequences of doing so results in serious social and environmental problems (Altieri 1996).

The repeated use of transgenic crops in an area may result in cumulative effects such as those resulting from the buildup of toxins in soils. For this reason, risk assessment studies not only have to be of an ecological nature in order to capture effects on ecosystem processes, but also of sufficient duration so that probable accumulative effects can be detected. The application of multiple diagnostic methods will provide the most sensitive and comprehensive assessment of the potential ecological impact of transgenic crops.

Although biotechnology is an important tool, at this point alternative solutions exist to address the problems that current GMCs, developed mostly by profit motives, are designed to solve. The dramatic positive effects of rotations, multiple cropping, and biological control on crop health, environmental quality and agricultural productivity have been confirmed repeatedly by scientific research. Biotechnology should be considered as one more tool that can be used, provided the ecological risks are investigated and deemed acceptable, in conjunction with a host of other approaches to move agriculture towards sustainability (NRC 1996).