You list five topics for comment:

1. GM food safety. I sense little real concern here. The existing mechanisms for evaluating 'novel additives' are never going to give absolute safety but they will evaluate obvious 'novelties'; conscious manipulation of food (including GM) will be tested in a way that there was no reason to do for (say) BSE. Even Prince Charles almost accepts this. In his 8 June 1998 Daily Telegraph article (which is his most detailed critique known to me), he excepts "certain highly beneficial and specific medical applications" from his general anathema of GM. The inference is that he has no blanket objection to GM. The obligation clearly is to carry out food safety testing in ways comparable to new drug testing. The requirement is to ensure that food safety testing is carried out in as open a way as possible. The debate about 'substantial equivalence' (Nature, 401: 525-6 and subsequent) is relevant but probably a red herring.

2. Gene flow and detection. Gene flow from GM organisms is going to happen unless they are made sterile. Does it matter scientifically? The objections from organic enthusiasts has more to do with philosophy (and perhaps legal definition, which may need amending) than scientific detriment. The substantive problem is whether unwanted genes will leak from GM organisms and cause unwanted effects. The recent report (Nature, 421: 462) that weeds crossed with GM plants can be less vigorous than 'normal' weeds is unsurprising. Repeatedly it has been found that intentional crosses into wild stock (of turkeys, red deer, angola rabbits, etc) yield inferior offspring which fail to survive or contribute significantly to future generations. The only possible exception to this known to me is the effect of farmed salmon escapes, but hard data here are lacking (I think - I may be wrong); the problem here is competition between feral and wild fish. Hazards from introductions of aliens due to changes in distribution (as a consequence of climate change) or chance or deliberate releases are much greater - witness Rhododendron ponticum, Japanese knotweed, Canadian pond-weed, mink, Sika deer, rats, etc.

A concern frequently expressed by anti-GM campaigners is that introduced genes may produce unexpected effects due to unforeseeable 'position effects'. Whilst this is true it is over-emphasized. One area of research which could be usefully summarized in the GM context is the extent of naturally occurring 'horizontal' gene transfer. The existence of this gives the lie to the common canard that movement of a gene from one species to another is wholly against 'the grain of nature'/ 'God's will', etc. Horizontal transfer has been of considerable importance in the evolution of bacteria and certainly happens in higher organisms, as shown by Barbara McClintock in maize and subsequently by others in Drosophila and mice. I have seen attempts to estimate the rate of incorporation of foreign DNA, but I have been unimpressed with them and hence made no note of the sources. However retrovirus biology is a topic of considerable importance (not least among cancer biologists), and there should be the possibility of finding/commissioning a review of horizontal gene transfer. Such a review would counter the accusation that GM is unnatural and therefore probably immoral, and would give positive support to understanding the 'natural' consequences of DNA introduced by means other than conventional sexual reproduction.

3. Environmental impact of GM crops. The main problem here is the risk to biodiversity, both by the possibility of targeted pest and weed control, and by the indirect effects on organisms higher in the food chain (such as seed-eating or insectivorous birds). This risk has certainly been over-emphasized in the context of GM because the whole trend in modern farming has been to minimise the occurrence of unwanted (weeds or 'volunteers'), thus creating as near approximations to monoculture as possible. There has been a cataclysmic decline in many farmland specialist bird species as a consequence. But this has nothing to do with GM, despite irresponsible scare stories of the dangers about the latter (Christian Aid produced a dreadful Report Selling Suicide in 1998 which purported to be about GM but was really about the stresses and problems of changing from subsistence to market agriculture). There is no intrinsic reason why GM will necessarily be more malign to biodiversity than current farming practices, but continued monitoring is clearly going to be important. BTO will carry on with their work (assuming they continue to receive contracts to do so); there is a need to increase the monitoring of invertebrates. This is currently very patchy. It could help considerably if the National Biodiversity Network gets its act together; perhaps someone could help by directing money towards greater effort on the national recording of key invertebrate groups.

One of the most interesting comments to me at the Royal Society meeting was Andrew Watkinson's response to a question as to whether a 'baseline' level of biodiversity should be laid down. His reply was that we had to decide upon an acceptable level of biodiversity and then aim our management towards maintaining (or achieving) that level. In other words, biodiversity is a subjective (anthropocentric) response, not some basic biological measure. This is certainly true with regard to farmland bird declines, because these have taken place in what is effectively an artificial environment - that of human agricultural systems; the species concerned reached their abundances because of a particular type of agriculture. It would in theory be possible to aim for a biodiversity characteristic of non-human influenced land, but this would be artificial, impossible to attain and hypocritical since we live in an irrevocably human changed system. But it points up the (non-scientific) fact that our perception of what 'ought to be' is determined by culture rather than biology. It leaves open, of course, whether there is a 'minimum' biodiversity for efficient functioning of natural systems. Increasingly we are becoming aware of the importance of 'nature's services' in maintaining carbon fixation, soil stability, watershed functioning, recycling of nutrients and pollutants, etc (see Costanza et al., Nature 387:254-60; Brown, L.R. Eco-Economy, Earthscan, 2001; etc), but this is to widen that debate beyond agricultural land. Nevertheless, these are issues which should not be forgotten in the context of the present Debate.

4. Future Developments. Bob May commented at the Royal Society meeting that his main concern about GM was the 'ratching-up' of the intensification of farming practices. The issues here are real because they raise again the spectre of BSE, although they are not directly related to GM. There is already debate about the legitimacy of 'industrialisation' of farming practices, going back at least to Ruth Harrison's Animal Machines (1964) and the subsequent Rogers Brambell guidelines for husbandry, followed much more recently by the Banner Report on the Ethical Implications of Emerging Technologies (1995). Scientists are already heavily involved in these issues (e.g. at the Babraham Institute) and care must be taken to ensure proper funding for this work. There has been a tendency over the last few decades to reduce finance for research not concerned with increasing productivity. But this does not directly influence decisions that will have to be taken about GM use.

5. Regulatory Process. There is an important educational exercise here. I am not thinking of conventional education, but the sort of misconception cited in the House of Lords Select Committee Report on Science and Society (2000), that the only genes there are, are those involved in GM. The HoL Report describes the processes leading to the setting up of the Human Fertilisation & Embryology Authority. I was a member of the HFEA for the first six years of it existence and was very impressed with its structure and modus operandi (in recent times it seems to have got bogged down in legal complications). It was (and is) a public and independent body, responsible to Parliament. Its Reports to the Secretary of State are public documents open to debate and challenge by anyone. It carried out a number of wide consultations that attracted some sensible comment (as well as the inevitable 'postcard campaigns' about particular topics). My impression is that the Food Standards Agency is designed to operate along similar lines. Is there a place for a GM Authority - perhaps along the lines of the old GMAG, which effectively did itself out of a job?

My comments may be way off target. In no way am I an expert on GM and I may have misunderstood some of the issues. In reality, the most problematic issues are those at the borders of science, where science meets society. Natural scientists have had a bad habit in the past of leaving such 'fringe' issues to social scientists. Most natural scientists are still have not convinced of the need to contribute more fully to such topics. I have worked for some time in the British Ecological Society to encourage debate on 'ethical' issues. There is no doubt that many ecologists see this as important, but are unsure about how to handle them (e.g. Trends in Ecology & Evolution, 14: 259-260). Some years ago, I chaired the Steering Committee of the Joint Agriculture and Environment Programme, set up by NERC, AFRC and ESRC in the light of heavy criticism of over use of pesticides and toxic chemicals in farming. It took two years hard work to convince the natural scientists involved that there was more to their work than merely producing hard data and an even more difficult task to persuade the social scientists that the sole function of natural science was not simply to produce data for their use. These barriers must be broken down if there is to be a trust in 'science' and a sensible debate about GM and its possibilities - particularly in the developing world.

Finally, public doubts about GM (as distinct from pressure group propaganda) are undoubtedly fuelled by the perception that GM is a mechanism for the big agrochemical companies to make profits, and that it has no benefit to the consumer. A counter to this is that we shall need increased productivity to feed an increasing [developing] world population, but objectors answer that the problem is poverty and distribution, not the amount of food that can be grown. Stuart Pimm (in his horribly titled The World According to Pimm, McGraw-Hill, 2001) shows the error in some of the calculations used, but I believe there would be value in producing a robust scientific analysis of food needs, population projections, and agricultural potential under various technologies (including 'traditional' and GM). This may have been done, but I do not know of any such analysis. Most accounts I have read are submerged by polemics and politics. The World Scientific Academies Report Transgenic Plants and World Agriculture (published by the Royal Society, 2000; <http://search.nap.edu/html/transgenic/notice.html>) is excellent as far as it goes, and the data I am requesting may be hidden in the base documents behind this Report. But I suggest that an analysis of population food needs and provision based on as hard data as possible could contribute positively to the GM debate(s) in the same way that the IPCC has swayed the understanding of climate change.

Professor R.J. Berry
Department of Biology
University College London
Gower Street
London. WC1E 6BT.

Professor R.J. Berry is an animal ecologist geneticist and was Professor Emeritus of Genetics at University College London 1978-2000, and a former President of the British Ecological Society, the European Ecological Federation, and the Mammal Society. Professor Berry is also Moderator of the Environmental Issues Network of Churches Together in Britain and Ireland (and a former President of Christians in Science).