WEDNESDAY 3 JUNE 1998

DR NIGEL J POOLE, DR DAVID A EVANS and DR SIMON W J BRIGHT

  80.  Should labelling go further than stating that a product is genetically modified, and state the purpose of the modification, as may be the case in Canada?
  (Dr Poole)  I think we need to come back to what a label means. The label gives information. It must be verifiable and true; otherwise there is no point in having a label. I think the way you are trying to take it forward is actually giving a need. I think it is part of the communication exercise of why you are selling the product. If you looked at our tomato puree it actually says there are environmental benefits. I think it is actually best left to the producer, the company, to explain why they have done it. I do not think that should be a mandatory reason.

  81.  My reasoning behind this is that the consumer might beware a label saying it is genetically modified. The next question is, "In what way is it genetically modified? Is this something I wish to purchase or should I be frightened off?" or whatever the case may be, and I just wondered whether that was something that could go on the labels, "genetically modified to——" and then tell you what the reason was.
  (Dr Poole)  I think that is a very good argument and there is a lot of logic in that, but the final decision should come down in the end to the producer. I do not think that it is a role of regulation to help us sell products. We have to sell products. I would question whether or not the label is a major turn-off for people. I am sorry: we keep on talking about tomatoes but it is the only product we have got which we have experience with in Europe. We are outselling the other product. Over the two years now that that has been on the market it has sold side by side with the other product and if you look at the cans it is very clearly labelled, "Produced from genetically modified tomatoes". I do not think it scares people. I think it is treated as giving people information and that must be right.

  82.  You see it more then as advertising rather than as information?
  (Dr Poole)  No, giving information is not advertising. This is giving information which I think we should do. We have not spent one penny on advertising, none of the companies has, on this product.

  83.  If labelling is to be required, should it be required only to the extent that the labelling can be verified by testing? What thresholds should be set?
  (Dr Poole)  I think that is a very important question. The label is so important. It has to be verified. You have got to be able to measure it in some way. Otherwise there is no point in having the label there. We, the regulators or the inspecting services will not be able to determine what it means unless you can actually measure it. I think that that is a very important area. Thresholds are also important here. We get back to the debate about understanding percentages, I suspect, and the mathematics. All through agriculture we are used to thresholds. When we sell hybrid seed, the accepted purity level of hybrid seed is, from memory, about 95 per cent for most hybrid seeds. If you take the organic label, it says that 95 per cent of the products must be organic. Those are their standards. Those are normal threshold standards which are used throughout the food industry and they have to be brought into our normal, legal regulatory system.

  84.  What lessons in allaying public concern have you derived from your experience of marketing the genetically modified tomato paste in the United Kingdom, and I ask this question particularly when I think public concern has been heightened by recent publicity, for instance with regard to maize modified to control corn borers and its effect on lacewings in laboratory experiments, and also the publicity which there has been about the effect on ladybirds with regard to genetically modified potatoes. All these things increase public concern but perhaps you could tell us what lessons you have learned in allaying that concern?
  (Dr Poole)  Let us take it down step by step. The most important requirement is that you have to have a product which is a good product that people want to buy. The second important requirement is to get that regulatory approval and, as we said in the written paper, one of the reasons we had to bring this product to the United Kingdom was that we had had a long history of novel food guidelines operating within this country which had been copied around the world, so you had that safety requirement. Thirdly, I think it is providing information, which is part of the label, it is showing respect for your customer, which I hope we did. Ours was the first United Kingdom product so we were setting the standards if you like. It was very important to us to actually respect the consumer, listen to the consumer, and provide that information. I think those were the major requirements. The other point which we find it very hard to deal with is this sort of soundbite way of looking at risk assessment which is going on. An orgy of "all GM foods are dangerous". As we say in our paper, there is no such thing as a generic or universal genetically modified crop. Every species, every variety, is different. Tomatoes are very different from maize. The traits that we put into our tomatoes are very different from the traits we put into maize or soya, so each one has to be done by a case-by-case system and analysed that way, and that is a very hard fact to communicate to society. We have got to be much better at communicating the diversity of agriculture and food.

  85.  Surely you are more likely to get unexpected effects, and I can think of one of provoking allergy reactions for instance, following genetic modification rather than plant breeding. Surely that is right. Do you not see danger for instance in genetically modified foods provoking allergic reactions?
  (Dr Evans)  Yes. Allergy towards proteins, and these plants are producing proteins by and large, is a well studied subject and there are tests available in toxicology laboratories which detect these, so it is tested in that sense. The proteins are tested and indeed approved, so there is a regulatory framework which ensures this. If I may return to a partly related question when you talked about beneficials, the effect on beneficials of any crop protection treatment, whether it be chemical or genetically based, is subject to regulatory approval. It is a major part of the regulatory package. Those hundreds of environmental scientists that we employ are largely in place to deal with the environmental effects of our products and we know full well that success in the market place and success in regulation will be very largely based on the safety that we bring forward with our products, and that is why we use biotechnology, to come forward with exclusive products in terms of safety to beneficials, growers and the public. The package on safety typically will account for up to 50 per cent of the spend on a crop protection product in bringing it from invention to market.

  86.  What about the danger of transplanted genes interacting with or combining with other genes to produce a toxic reaction? Is that a real danger? It certainly is suggested as such by some.
  (Dr Bright)  I think that is a good example where again you have to turn it into a positive question that you are asking rather than saying, "Can you show that nothing has happened anywhere?" which is trying to prove a negative. It seems to me that for each gene you have then to say, "What are the anticipated effects?" and, "Do you see those?", and in going through the laboratory, glasshouse, small-scale field trials, you then have to say, "Is this behaving in the same way as it would going through a normal breeding programme?" You will have the same scrutiny there for unexpected effects in the environment, and then, if it is a food, when you are tasting it you will be asking questions about, "Does it taste like a tomato?", if it started out as a potato, "Does it still taste like a potato?" "Are there any other off- flavours?" and so on. Your testing process has to be there to ask questions about, "Is this still the same base? Has it still got the same hundred thousand genes that make it a tomato?" plus your one or two extra genes in there. You are always asking that against the standards, comparing it against the standards, it seems to me, rather than saying, "Is there a new interaction in general?", which is a very open-ended question to which you cannot get a satisfactory answer.

  87.  Is the furthering of antibiotic resistance through genetically modified foods a genuine cause for concern? How can the least-safe practices be identified and phased out?
  (Dr Poole)  Again there are only a very few antibiotics which are being used in plant breeding. We have used kanomycin resistance as a marker in some of our plants. We believe that that is a safe product, and I think that has been accepted virtually all round the world now by regulators and independent scientists. However, we have to accept the fact that people are genuinely concerned about antibiotic resistance. We are therefore moving very fast away from this sort of marker, but agriculture is a slow process and it will take some time before that happens.

  88.  What about the question on insect resistance? I notice in members' papers that areas are to be set aside called "refugia" areas where the crop has not been treated so that insects can survive on these crops. Am I right in thinking that sprays are never 100 per cent anyway and they only reduce insect populations? Is there a likelihood that these refugia areas could lead to insect resistance being built up?
  (Dr Evans)  The purpose of refugia in a crop which is expressing an insecticidal agent protein is to ensure that beneficials have a breeding ground on that crop, so it is in fact performed for the exact reason of ensuring that we do not get resistance to that particular insecticide expressed with the crop. In terms of the development of resistance to insecticide, we know that continuous pressure on the insect of the agent can cause resistance. That is true of chemicals and genes and that is a well studied phenomenon. The refugia is set there to avoid that problem. Levels of safety are built into that and in some countries of the world the percentage of refugia is quite large. As you probably know, in Australia it is a large refugia. Essentially we are doing an experiment there but it is an experiment which has got a huge safety margin. The purpose of doing that is to ensure that the effect is one to which resistance does not develop and so that effect can be used into the future.

  89.  What is the danger of genetically modified genes escaping to weed relatives and producing a superweed that presents a control problem? Can the risk be defined or quantified?
  (Dr Evans)  The concept of a superweed is very interesting and quite difficult to grasp. We have all seen The Day of the Triffids and I guess that can cause some alarm, but frankly I do not believe it is a problem in that farmers are used to dealing with escapes and volunteers, weeds they do not want in their crop, by using herbicides. Provided that there are enough selective herbicides on the market which kill weeds or take care of these volunteers (which is the phrase we use), that will not be a problem. Presently I can say that there are literally very many more than 10 different types of action of herbicides, so a herbicide will be available for control. In terms of the modelling of this, it has been modelled and indeed it will happen and we know the extent to which it will happen, but the control will not be the problem.
  (Dr Bright)  Can I just add to that, that the rate of outcrossing if you like between crops, weeds and feral species is different for different crops. You have to do it on a case by case basis. Genes are very unlikely to travel out from a tomato for instance because it is largely self-pollinating. Other crops have different breeding systems and the birds and bees do what they are going to do anyway in that crop. You also have to look at it case by case by what the gene is. If a gene is conferring resistance to a particular herbicide application, then that is going to have a different effect on selective advantage of weed that receives that or a plant that is left behind in the following year. If it is a different trait, if it is a trait for changing starch content, it is very unlikely to have any effect on selective advantage. It is certainly jolly difficult to become a superweed.

  90.  None the less the fact that there could be this outcrossing from pollination could lead potentially to a need for a farmer to have more chemicals in his armoury rather than fewer?
  (Dr Bright)  We think it is a zero sum. Again, there is a strong incentive to control that and to make sure that you have thought about that, and that is part of the regulatory assessment that goes into the release in the first place. We think it is actually quite neutral. There is always a requirement by farmers for better means of control and weeds are always evolving resistance to natural gene based resistances or chemical based resistances. It is a new example of an old problem if you like which has been going on for quite a while.

  91.  Would you like to comment on Greenpeace's call to ban the planting of rapeseed because of the outcrossing dangers associated with it?
  (Dr Poole)  We are not actually involved in that area, so it is difficult for us as a company to give that comment.

  92.  I detect that quite a body of opinion is to the effect that they would prefer their crops not to be sprayed rather than to be sprayed with something to survive a poison. Would you say that herbicide-tolerant crops would encourage more damaging use of herbicides in the long term?
  (Dr Evans)  I think the statement that chemicals are seen as poisons actually has to be examined. The current research in the chemical part of crop protection is dedicated towards producing superlative chemicals which are extremely safe. That is what the regulations require and that is what the companies are producing. It is possible to invent a herbicide, and I can think of two which are particularly used at the moment, which act on biochemical systems which do not hurt human beings, and these molecules have remarkably low toxicology. One of these in particular is the basis of herbicide resistance to crop technology at the moment. It does not follow that chemicals, either because of their widespread use or even because of their inherent activity, are more toxic to other species. There is no general rule in this case by case. The point I am making is that the chemicals used in that context are extremely safe, so I do not think the chemical/poison theme stands up. Our strategy as a company is to provide the farmer and grower, and indeed the stakeholder, with the optimal combination of safe chemicals and safe genetic technology. We see it as an integrated future in that context.
  (Dr Bright)  I have been involved in herbicide resistance research in terms of saying, "What can you do?", I guess for the last 15 years, and it has actually come down in the end to saying that with all this wonderful technology actually the regulation is around the herbicide. If it is a good herbicide and you can build in selectivity or change selectivity through genetics, then that is fine. If it is a bad herbicide this does not make it any better. It is really all about how we regulate the chemicals and those are very highly regulated and the hurdles are getting higher. In that sense it does not follow that you can alter selectivity either through finding a new chemical which has that selectivity or by building it in through selective breeding or through biotechnology. It is essentially about changing selectivity and that is one of the only examples. My personal preference is for spraying weeds to kill them rather than having to dig them up one by one but other people might have a different choice and as long as the choice is there it seems to me that is fine.

  93.  Could I ask whether your marketing strategy is moving in the direction that some other companies have of very closely linking a genetically modified strain with a particular own brand herbicide and marketing the two as a package?
  (Dr Evans)  Certainly bundles on a crop are attractive packages to farmers and growers. It is impossible actually to tie up a whole market; there is not that amount of trust. There is always choice, and even in the cases of the bundles the farmers and growers have an extremely wide variety of choice in all those crops. There is one marketing strategy to provide the farmer with convenient bundles of both gene and chemical effects on particular crops.

  94.  Can I declare an interest? My wife has a small hill-farm in Wales. As you know, the draft revision of directive 90/220 proposes that genetically modified crops should be monitored for environmental impact after commercial approval has been given. Do you think that is right and do you think it is feasible and desirable that there should be such monitoring?
  (Dr Poole)  This worries me because it is again that soundbite answer to science. This is the solution to somebody's concerns: monitor. Several times in our messages when we give them we have to come back to the science basis. The most important thing as scientists you have drummed in to you time and time again is to define the question. Monitoring is fine if the regulators can actually define a question. What are we meant to be monitoring? The hill farm in Wales: is it feasible to monitor it? Otherwise you will be spending an awful lot of money achieving nothing. There can be nothing wrong with monitoring—it is like motherhood—but you must define the question and the rationale of doing it first. For example, with certain products in America they have said they would like a review after a number of years because of this insect resistance question. That is normal practice and there is nothing difficult in doing that and our legislation does that anyway at the moment.

  95.  So do you think the draft revision of the directive goes too far?
  (Dr Poole)  I think it is rather badly drafted at the moment. It is sloppy drafting.

  96.  Could I perhaps help Dr Poole with a question? If you were asked not only as part of the regulatory process in terms of the information you were going to provide at various testing points but also as part of the post-release monitoring process to monitor not only the impact on beneficials but also the impact on other plants and particularly the broad land use pattern that is resulting in the vicinity from changed decisions by farmers about how they manage their land, would that be reasonable or unreasonable to ask you as part of that process?
  (Dr Poole)  Again, this is not the area to go down without a scientific based question, but I think it is legitimate to ask us are we changing the beneficial insects, do we see a use pattern change in that? That is a very fair question and we would be look at that question anyway.
  (Dr Bright)  Only if you had an insect resistance———
  (Dr Poole)  I am assuming that.
  (Dr Bright)  If you had a crop where you had changed the quality for industrial use and it seems to me you would have to ask the question about beneficials if that was one of the things that you expected to change.
  (Dr Poole)  Let us come back. It is defining the question for the crop. You then asked about land use and other take-up of the technology. I am not sure that is our responsibility. I am not quite sure how we would measure it. You are looking, as perhaps English Nature (Or is it the Government's) concern at how technology is being used. We would report how our products are used but determining the sociological aspects I think would worry me even more. That is the sort of question we would look at. You would have to define before you started the questions you examined and the answer you wanted, which is just as important.
  (Dr Evans)  One of the commonly used phrases is that association does not mean causation, so it has to be a question that is framed scientifically properly because expending our effort chasing the wrong question which is not properly defined is a waste of resources and can find totally inappropriate solutions. You have to be able to come down to define a sensible scientific question which can be scientifically answered.

  97.  In the long run could you tell us how viable you think is the segregation of commodity crops? Could you also tell us whether you think there is a role for governments in requiring segregation or is it a matter best left to the market?
  (Dr Poole)  If you look through our research portfolio, a lot of it is into quality traits. The only way we can get value back to pay for research is segregation of those products. That is the only way that can happen. If we come to segregation of commodity crops, I believe if there is a real market pull, segregation will happen along the whole of the food chain. The agri-food chain is a very complex food chain but there has to be evidence of the real market pull for it to happen. At the moment I am not quite sure on the commodity crops but in the future we see segregation coming in as part of the normal agri-food chain.

  98.  What about the Government role?
  (Dr Poole)  I think in the end it has to come down to the market deciding that segregation is needed. The government role is to decide whether or not they wish to have these crops in their particular country.
  (Dr Bright)  Once they are approved then they should not have that problem; they should say, "These are now normal products" and that seems to me to be the key thing that the regulation system has to provide, to say, "Yes, these are equivalent" and if they are not equivalent then you are required that they need to be kept separate or whatever. If you are really saying they are a commodity, wheat is wheat or whatever, that probably should be the end of it. You should not be discriminating.

  99.  Could I ask you a final question on which you have not had notice? Does your company take out insurance policies to protect itself against the possible consequences of unforeseeable consequences of genetic modification?
  (Dr Evans)  I cannot answer that question. I do not know the answer.
  (Dr Poole)  My belief is that as a company we carry our own insurance for those areas and the answer on liability which Dr Evans gave much earlier is part of that answer, but I will check with the lawyers.
  (Dr Evans)  We certainly have insurance policies against failures in crop and so on, so we have insurance, but what I cannot answer is if it is specific in relation to your question.

Chairman]  Thank you. The Committee is grateful to you for having come to give evidence. We have had a fairly brisk canter over quite a long haul. It may be that we have broken into a gallop at some stages. If you feel that you have not had time to answer any of these questions fully, we would welcome any amplification in written form which you wanted to send in. We are extremely grateful to you for coming and giving us your point of view on this subject. Thank you very much indeed.