From there began the modern era of biotechnology, made possible by advances in in vitro cell culture and the use of molecular techniques both to identify and - in everyday computer language - to "cut and paste" genes from one cell to another. This process of recombining DNA is known as genetic engineering and its products - microorganisms, plants, trees, livestock and fish - are called genetically modified organisms (GMOs). At a pace paralleled only by the digital revolution, we are now unravelling the structure of genomes, discovering the functions of individual genes, and understanding how these can be turned on and off to change the traits of whole organisms.
Analysing the benefits. What benefits have the current GM crops brought to those who develop them, grow them and consume their products? The picture is still unclear due to the paucity of large-scale monitoring and economic studies. US Department of Agriculture data indicate that although there has been no significant reduction in the total use of herbicides, farmers have switched from using three or four different compounds to using only one or two (the main one being glyphosate, which is generally recognized to be ecologically less harmful than many of the other herbicides that it has replaced). One recent study estimated that GM varieties helped US soybean farmers save $216 million on weed control in 1999.
Regarding insecticide use following the introduction of "Bt crops" (containing genes from Bacillus thuringiensis, or Bt, a soil bacterium that produces a protein toxic to certain insects), the picture is somewhat clearer. Estimates of savings vary, but one study puts the overall reduction in pesticide use in the USA at 1,200 tonnes per year, suggesting significant environmental benefits.
In developing countries, the picture is fuzzier. Since farmers in Argentina's Pampa Ondulada started growing GM soybean, glyphosate has taken over from what was formerly a package of herbicides, and its use has tripled. Interestingly, both in Argentina and the USA this crop is grown in rotation with wheat, maize or sunflower and has promoted the uptake of reduced-tillage practices.
In the absence of empirical evidence to the contrary, it is fair to conclude that, to date, the main beneficiaries of GM technology have been farmers and technology providers. Also, the main effects noted are reduced use of insecticides and a switch to more environmentally friendly herbicides. There is, however, little to indicate that yields are significantly enhanced compared with conventional practices.
Assessing the risks. But what about the potential risks of GM crops to the environment and human health? The main environmental concerns include transgene spread to other crop varieties, wild relatives and weeds, and to target and non-target insects, viruses and fungi. On the health side, possible risks are seen as the emergence of new allergens and of bacteria resistant to antibiotics.
Some of these concerns are real and cannot be ignored. The developers of GM products should concede that including antibiotic resistance genes is hardly conducive to improving public acceptance of the technology. On another front, the problem of pests developing resistance to control measures - the bane of farmers for decades - means that carefully designed crops and resistance plans are essential. But even, for example, when high-dose Bt rice plants become available, the institution and enforcement of effective refuge systems in subsistence systems may prove difficult.
Another issue: while genetic engineering offers great specificity in terms of what is put in, it has - as yet - limited control over where the construct lodges in the genome and therefore over how the new combination of genes will behave. This raises the possibility of genetic and phenotypic variability. Although this should be detected in crops during laboratory and pre-release trials, in the case of trees detection could be more difficult due to their much longer life spans.
Genes do move in ecosystems, from one crop to another crop of the same species and to closely related wild species and insects, weeds and soil bacteria. This occurs with conventionally bred crops and GM crops are no exception. The issue, then, is not whether gene transfer can happen. It is whether a GM crop presents a different kind of hazard and a different level of exposure under practical farming conditions compared to the same crop bred for the same phenotypic characteristic using conventional breeding. In other words, does it matter?
Up to now, relatively little investment has gone into assessing the possible longer term environmental effects of GM crops. However, so far nothing negative has been found. Late last year, the European Commission's External Advisory Groups gave these crops and the foods derived from them a "clean bill of health", reporting that extensive research had not come up with any new risks to human health or the environment beyond the usual uncertainties of conventional plant breeding. The US National Academy of Sciences came to essentially the same conclusion in 2000. Nevertheless, both reports called for clear regulatory oversight and greater attention to improving the public's understanding of the process and principles that underpin it.
Where to now... Biotechnology is big business in the industrialized world. Although the lack of perceived benefits for consumers and concerns about food and environmental safety have slowed adoption of GM products in some countries, its seemingly limitless potential for future application, paralleled by the strengthened and wider outreach of intellectual property regimes, has tilted the balance of investments significantly towards the private sector and thereby towards the global commodity market.
In some developing countries, large investments are being made in biotechnology and GMOs, but neither the techniques nor the products are being directed sufficiently strongly at the needs of low-income farmers. The likelihood, therefore, is that it will be the larger, more commercially oriented farmers with an eye on higher value export commodities who will gain most. Other beneficiaries will be private seed companies and the shareholders in multinational corporations.
If biotechnology, including GMOs, is to deliver on its promise of alleviating hunger and poverty in the developing world, it must target the crops used and the traits needed by the poor, including greater emphasis on traits for quality and resistance to abiotic stresses. Further critical needs are to loosen up the arrangements for enabling developing countries to access proprietary technology, and to provide poor farmers with improved seeds while protecting them against inappropriate restrictions in propagating their crops. The initiative to develop nutritionally reinforced rice - and the decision by the holders of the related patents to make this product widely available to developing countries without claiming intellectual property rights - are positive steps in this direction.
Imported crop varieties or gene constructs will not, in themselves, lead to sustained improvements in productivity or diversification into higher value systems of agricultural production in developing countries. Only local breeding, allied with systems-based approaches, can address local needs and cultural preferences and provide sufficient, safe and nutritious food. The same will hold in the livestock, fisheries and forestry sectors. This is true now and it will remain so in the future as the functions of more genes are determined and become candidates for transfer.
Like any tool, biotechnology will not work any magic on its own. Success in its deployment requires careful identification of the opportunity or need to bring it into the wider technological package being considered. It also requires careful assessment of the costs and risks involved and the benefits that are likely to accrue. And just as importantly, these decisions need to be made within a policy framework that promotes and protects the public interest.