Official Report 199KB pdf
We now move to agenda item 3. We have David Robinson, Gavin Ramsay and Geoff Squire with us this morning. We try to keep these events as informal as we can, but as we seek information from you we will go through the process of asking questions.
I will say a few words to begin with. As you will have seen in my biography, I was a member of the Advisory Committee on Releases to the Environment from 1990 until June last year. That gives the perspective with which I will look at the questions that we consider today. I should point out that the time that I was on ACRE was pre-devolution. I cannot comment much on the extent to which things may have changed in the past 18 months or so, but I suspect that they have changed little.
Thank you. Your paper provided a useful overview.
Dr Robinson, thank you for your paper, which is fairly clear on the relationship between us as politicians and you as scientists. We have responsibilities as well, and ultimately the value judgments have to come from us. You make the recommendations and give us the science and the facts, but we live in a democracy where responsibility for making judgments does and should remain with politicians.
There is very little satisfactory evidence. There is some evidence that in certain circumstances genes can be forced to flow into some of the organisms that you mentioned, but there is no clear evidence that the process is likely to occur in nature. My colleague Dr Squire may be better qualified to answer the question.
In general terms, Dr Robinson has given a fair answer. We can force exchanges and cause things to happen, but the soil community is very resistant to influx. DNA can be found in soil in various forms—we can extract soil DNA and RNA. That is part of the genes of different organisms and it is difficult to work out where it comes from. This is a difficult area, but I support the general statement that Dr Robinson has made.
Are you aware of any research that is being done at the moment in this area?
Not in our organisation. There is little research being done generally in this area. I would need to check, but the UK Government has not put much money into it. At the moment, it is not a very active area of research.
Would you agree that, if we proceed to plant GM crops on many acres of ground, there is a long-term possibility that changes of which we are unaware will take place in organisms in the soil?
One can never exclude every eventuality and it would not be right for me to say that it was not possible for a certain kind of biological event to happen. However, there are many other things that might and might not happen. At the moment, most independent scientists would say that gene flow of this sort was on the remoter end of possibilities.
There are always two questions: whether it is going to happen and whether it matters if it happens. If there is a possibility that the introduced genes from GM crops can move into soil organisms, that implies that DNA from plants has been moving into soil organisms all the time. There is no obvious reason to believe that DNA coming from GM crops is any more significant than plant DNA that must have been moving all the time.
I accept that, for the moment. Can you expand on what you still consider the major risks from field trials of GM crops?
I do not think that we have yet identified any major risks. There are a number of hazards. Ought I to justify the difference between a hazard and a risk?
Yes please.
A hazard is a characteristic of the organism that might lead to a scenario in which harm might occur. When a hazard is identified, a risk assessment is carried out, which defines the probability of that scenario occurring. A hazard is a hypothetical way in which harm could occur, whereas a risk is the actual possibility of that harm happening.
Can you expand on the hazards that are being investigated in the Scottish field trials?
I shall pass that question to Dr Squire, who is more closely associated with those trials.
The principal matter to be examined in the UK-wide trials, some of which are being held in Scotland, is the potential effect of a GM crop that is resistant to an effective herbicide being grown in conjunction with that herbicide. In that case, the question is whether that combination will be so effective in suppressing the arable flora—the weeds—that it will lead to degradation of the diversity in the field. That will have knock-on effects on the organisms that eat the weeds and the other organisms that prey on those. It is primarily a question of whether the GM herbicide-tolerant crop is so effective agronomically that it will lead to a decline in the biodiversity. Currently, weed control is generally effective but often haphazard. To make it really effective, persistent and highly toxic chemicals must be used.
Thanks very much. My colleagues may have one or two detailed questions on that later, but you have made that issue clear. For my final question, I invite you to reflect on the possible benefits of GM crops—the ones that we are experimenting with at the moment.
I stress that our role is that of neutral observer and arbiter. The potential benefit of this specific GM crop is that it gives the farmer increased flexibility. Often, herbicides are sprayed on crops fairly early in the season as a kind of insurance when farmers expect a weed problem. Weeds are persistent; many are quite damaging, but one can clear completely a field of weeds, if one wishes. The supposed benefit of this herbicide-tolerant crop is to give the farmer the flexibility to decide later whether to spray. He knows then that he can kill the weeds if he sprays. The herbicide is not persistent beyond contact with the plants and soil, which means that timing can be quite specific and that any vegetation that germinates after spraying will not be affected by it. We are trying to find out whether that will prove to be a benefit that will outweigh the potential disadvantages.
Will that result necessarily in less spray being used?
At this stage in the experiment, it is difficult to say whether less or more spray would be used. A different kind of spray might result that would be less persistent in the environment.
That is quite an important point. You used the phrase "less persistent". Are sprays in use that are more persistent in the environment?
Yes.
Are they less effective in killing weeds?
That is correct. However, the difference is that this particular GM crop is tolerant to a herbicide that kills virtually all weeds, but begins to do so immediately through contact. The actual death occurs a while later. Although other types of herbicide might be less effective against all types of weed, they are more persistent and hang around for longer.
I should apologise for the noise, which is because of the University of Edinburgh's contractors mending a roof, not our contractors mending our roof. We are trying to deal with the situation, but I am afraid that we will have to tolerate the noise in the meantime. Are there any other questions on the matter that Robin Harper is investigating?
Do you accept that the risks that the public perceive and that others might perceive in the sale of Scottish produce abroad arise solely from a scientific perspective and that, from that perspective alone, we cannot take risks?
As a scientist, I tend to view the matter from a scientific perspective, but I suppose that there are economic risks and so on. Furthermore, people believe that certain things are risks—things that I, as a scientist, would call imaginary risks. However, such risks exist.
I am glad that I studied risk assessment in the first year of my MBA and can understand most of this.
It has been said that GM crops are sterile, which minimises the chances of cross-fertilisation and the environmental risks that they present. However, when the Rural Affairs Committee took evidence on the subject, Simon Cooper of the Scottish Agricultural Science Agency stated:
GM crops are not necessarily sterile. Indeed, as far as I am aware, those in the field scale trials are not sterile.
I would like to reinforce what David Robinson has said: most GM crops behave as normal plants and are, in effect, normal plants. The sterility applies only to the contaminant within the seeds that Advanta sold. That arose as a result of the means by which hybrid seeds were made. That was done using a special system that relies on male sterility, so when Advanta's seed production fields were contaminated with pollen from an unexpected source—in this, case, a GM source—the seeds that Advanta had mixed with that stock were, when they grew into plants, unable to make pollen. They were not sterile—they were still fertile on the female side—so they will contribute to seeds being made by those crops.
The second part of the question was about the consequences of pollen from genetically modified crops spreading. In the case of oil-seed rape, pollen from GM crops can pollinate other crops—for example, feral rape and a couple of weed species, such as wild cabbage. The likely effect of that would be that the introduced gene would spread to some extent in wild populations of oil-seed rape. That would be limited to a very small number of species, but the gene would be likely to spread.
Oil-seed rape on roadsides is able to cross-fertilise freely with fields of oil-seed rape, so that is a route for genes to move out of fields and into semi-wild populations. Where truly wild species are growing in natural habitats, there is a theoretical possibility that hybrids will appear. If that happened, however, it would happen at very low frequencies. People have not observed such events with normal rape and are therefore unlikely to observe it with GM rape.
I would like to pursue that a little further. You said that there was "a theoretical possibility" of rape cross-fertilising with charlock, which grows in Scotland. I do not see that as an incredible danger; I see it more as an inconvenience for farmers and something that would undermine the purpose of developing a resistant form of oil-seed rape. Would it be possible for charlock to spread within oil-seed rape fields so that, in the end, the farmer was unable to kill the charlock—because it was also resistant to the herbicide—and he found himself harvesting charlock and oil-seed rape at the end of the season.
Yes, the situation that you have outlined is realistic. Perhaps in the long term it is conceivable that genes could move into charlock. They may take a long time to do so.
How long—10 or 20 years?
I cannot put a figure on it.
That situation ought to be dealt with by proper agricultural practices such as rotation. The farmer will not be growing the same crop in the same field every year.
I want to probe the issue of cross-pollination further and to consider exclusion zones. To what extent do you think that the exclusion zones around GM crops can reduce environmental risk?
They can reduce the environmental risk. Let us stick with oil-seed rape. Away from a crop, the amount of pollen from that crop falls off very quickly initially and then very slowly—it tails off. The further away one is from the crop, the less pollen is available from it. The pollination separation distances that have been adopted until now are not designed to prevent cross-pollination—that is a common misconception—but to minimise any harm from cross-pollination.
One cannot set an upper limit. There is a long tail-off, following the rapid decline over the initial tens of metres. That tail-off will take a very long time to fall away to zero and a limit cannot be put on it.
In your opinion, is 100 m adequate for the purposes that we are considering? Is it excessive, or should it be extended?
That figure is appropriate if you are willing to tolerate a low level of cross-pollination, such as one in 1,000 seeds or less, resulting from pollination between nearby fields.
Taking that point a little further, one in 1,000 does not sound like much, but it could mean a few thousand seeds.
We do not need an MBA for that one.
One in 1,000 seeds is a lot of seeds out of a field—it is 0.1 per cent of the product. We are talking about material potentially going into the human food chain, so 0.1 per cent of that material would be going into the human food chain. The judgment that must be made is whether that matters.
What about build-up in the local environment? There is no rotation in the local environment.
Are you talking about shed seed?
Yes.
Do you wish to comment on the significance of shed seed, Gavin?
Populations of oil-seed rape, for example, growing from shed seed, will make a contribution to the pollination occurring within fields, but it will be greatly diluted. In that case, pollen will be diluted to a greater extent than that from a nearby GM field, simply because a small population will not make a large amount of pollen. Shed seeds are important, as they could act as a reservoir of material that could provide GM pollen for future pollination, but the contribution to the level of contamination in crops will be minor.
What distance of exclusion zone is used in other countries?
I do not know offhand. I suspect that the distances in Europe will be fairly similar to those in the UK, but I am not sure whether there is one in the US at all.
Do you know where we would be able to get that information?
Probably with a bit of digging around it would be possible to come up with it.
We can pursue that issue later. As there are no other questions on exclusion zones, let us move on to the next subject.
I have a couple of questions on concerns about biodiversity, particularly about the potential effects on insects and other wildlife. Given the concerns about the possibility of pollen from GM crops ending up in honey or other products in the food chain, what risks do you associate with insects feeding on GM trial crops in the UK?
All the lab-based evidence on the present trial crops, in which genetic modification confers herbicide tolerance, suggests that there is no direct toxic effect on insects eating those plants. The trials are studying the indirect effects on the type, range and quality of the food that the insects eat and the knock-on effects down the food chain to other organisms, ending up with birds. As far as we are aware, there is no direct toxic effect, but we are examining the indirect effects of a reduced food supply for the insects.
People who assume that insects are pests may take a different view when it comes to the insects and wildlife that are perceived as beneficial to agricultural biodiversity.
That is absolutely right. If any kind of plant is affecting a pest insect, we must consider how it will affect non-pest insects and insects that eat the pests. It has been shown that there are potential effects.
You mentioned lab-based tests that have taken place. How do you think the risks should be assessed overall? You mentioned the difficulty of value judgments of what is harmful, but how can the risks really be assessed?
The difficulty with lab-based work is that the plant tends to be put in some kind of closed chamber with the pests and other organisms. The research monitors the way in which the pest eats the plant and becomes ill—or whatever—and the way in which the other organisms eat either the plant or the pest. The difficulty of extending that work to real life is that, in reality, those organisms have a choice and are not forced to eat the sick pest; they will find something else to eat.
Have the risks been fully taken into account by the UK Government's advisory framework on GMOs? Have all the issues been addressed adequately?
I am happy that, for all the releases that have happened to date, the risks have been properly considered. It is very important that things are considered case by case. I cannot comment on what might be proposed next week. However, from what I have seen I am happy that proper risk evaluations have been carried out on all the trials that have taken place up to now.
Good morning, gentlemen. I want to ask some questions about the Government's advisory framework on GM technology.
There are certainly regional differences throughout the UK, both in the environment and in farming practice. However, there is nothing that could be considered specifically Scottish. Different parts of Scotland are different from each other and the south of Scotland is very similar to the north of England. There are Scottish characteristics; for example, spring-sown oilseed rape is grown predominantly in Scotland and the north of England. Autumn-sown oilseed rape is grown further south.
When you talk about considering things on a case-by-case basis, you do not mean the GM crop but the GM crop in conjunction with the environment in which it is going to be grown.
The individual release is a case.
Do you consider the current UK Government advisory framework on GM biotechnology to be adequate? Can you expand a little on the strengths and weaknesses of the current framework? That may not be a very fair question.
It is a very big question. The current framework is working quite well. Whenever a number of overlapping areas are served by different advisory committees, there are potential difficulties. It is always important to ensure that there is an overlap and not a small gap. Over some years, there has been concern that there might be gaps. I understand that that is one of the reasons for setting up the new Agriculture and Environment Biotechnology Commission, which, among other things, is supposed to take an overall view and to try to ensure that there are no gaps. I would say that I am happy with the framework, as long as the AEBC—which has only just started its work—works as it is intended to.
Can you give us a run-down on the procedures that have to be followed before consent is issued for a GM crop trial in Scotland?
Yes. The proposer begins by producing a dossier that includes facts about the recipient crop, the way in which it has been modified, where the release will take place, the characteristics of the site, and exactly what they intend to do—when they intend to plough the land, when they plan to sow the crop and what they will do with the site afterwards. There is a list of more than 60 questions that they have to answer to supply information. There is also a risk assessment and a number of accessory papers in the dossier.
Are the case officers scientists or civil servants?
They are civil servants who have scientific qualifications. Most of them have PhDs in a relevant subject.
Do the dossiers prepared by applicants have to be supported by research findings and scientific evidence?
Yes.
Are they cross-referenced? Is there a mechanism for checking up on them, or are they taken as presented? If so, is that because they are previously subjected to peer review?
A dossier will be supported by scientific data, which might be a mixture of published information and the applicant's own data which may not yet have been published. The information is not exactly taken at face value. Part of the job of case officers, and more particularly of ACRE, is to scrutinise the evidence and look for errors. I can remember occasions on which a particular piece of evidence has been sent back and we have said that we do not believe the interpretation and asked for it to be done again. There is a to-and-fro process during which questions are asked and clarification is sought—evidence is definitely not necessarily accepted at face value. That does not mean that errors cannot be missed—we are all human—but the information is looked at in some detail by people who know about the subject.
Finally, do you consider there to be a role for the general public in deciding whether individual GM crop trials should go ahead?
There is an opportunity for public input into the system that I have just described. When the applicant sends his application to the DETR, he must advertise in the local newspaper the fact that he is making a proposal and give a phone number on which people can get answers to their questions. All the material in the dossier is potentially available to the public, except for certain bits that might be commercially confidential or that can be withheld for intellectual property reasons. However, the amount of detailed information that can be withheld is relatively small. The public can get the information and can make representations.
I want to ask a supplementary on that before I move on to my own line of questioning. There is a suggestion that the way forward for Scottish farming produce is to market itself as a brand of high-quality, top-of-the-range, pure and natural products. Do you accept that there is a risk that GM crops, or even the perception that there could be contamination, could damage that industry?
That is not a scientific risk; it is an economic risk or a risk to the public perception. It might equally be argued that by not embracing GM crops, the Scottish farming industry would reject some potential future advantage. What the scientists hope to do is to tell people to what extent GM crops are an advantage or a disadvantage.
Do you think it is legitimate for member states to be able to impose a complete ban on the cultivation of GM crops?
That is a political question. The only comment that I can make, as a scientist, is that, at least in mainland Europe, it would be very difficult to enforce such a ban because biology does not take any notice of national frontiers. There is a water barrier between the UK and the rest of Europe, although not between Scotland and England. I suspect that a ban would be relatively futile in the long run. However, it is not really a scientific question.
To some extent you answered in your comments to Nora Radcliffe my second question, on whether Scottish interests are adequately taken into account when decisions to permit a release are taken in Europe. You mentioned some of the procedures and the people involved. Who appoints those advisers? To whom are they accountable? Does anyone in Scotland have an ultimate right to veto a release?
Before devolution, when I was appointed, the members of ACRE were appointed as individuals by ministers—both English ministers and the Secretary of State for Scotland were involved in the process. We are now post devolution and post Nolan and the situation is more open. I believe that the ACRE posts are advertised and, in principle, anyone can apply. However, because it is an expert committee, there is no point appointing anyone who is not an expert. On the other hand, the assessors are officials who are appointed directly by the relevant minister and are presumably answerable to that minister.
The argument for an ability to opt out of a directive is that specific evidence could show that a trial might be harmful to the environment. This question might require a political or legal answer: would the definition of environment be restricted to scientific matters, in terms of risk, or could it be extended to economic risk, as has been suggested?
As I understand the legislation, harm to the environment does not include economic risk. However, I am not a lawyer.
Let us round up the evidence with specific questions on the Advanta situation.
What level of environmental risk is associated with the recent accidental contamination of conventional crops? What more could have been done to minimise the environmental risks associated with that accidental contamination?
Gavin Ramsay is probably better qualified to answer that question than I am.
As we have discussed, it is hard to envisage a real environmental risk in the harm that could be caused by the Advanta situation. It was an uncontrolled spread of GM material around the countryside and caused much concern. In that respect, it is disappointing that such an event took place. Companies such as Advanta would prefer greater clarity about the standards that are required for purity. As we have said, it is hard to achieve total purity in an environment where there are fields containing GM crops. Advanta clearly had some problems and cross-pollination occurred over longer distances than many people thought was likely. However, I doubt that there was damage to the environment as a result of that release. It is hard to imagine what real harm to the ecosystem could have resulted from that.
I have two final questions; one practical, the other scientific. I now understand the purpose of the trials. I have been given to understand that, apart from those, you do not have the resources to monitor pollen drift. Is that the case?
Organisations such as ours formed a consortium and bid for a contract to examine the ecological effects within the field and immediately around it. Since we were awarded our contract, a contract was put out to measure gene flow and pollen movement. We put in a bid for it but were not successful. The Central Science Laboratory was given the contract and its role is to consider pollen movement and gene flow away from those trials. I have not seen the specification of that contract and I cannot comment definitively on whether the resources are sufficient. We can follow gene flow within the fields, through the persistence of seed and so on.
So if we wanted to know how much pollen drift research is going on, we would have to write to the Central Science Laboratory, asking it what it is doing and when?
Yes. We know what it is doing, but you would have to do that to get the definitive answer.
My final question concerns resistance development. Bacteria and insects develop resistance to bactericides and insecticides. In the animal kingdom, the process of developing resistance is considerably longer. Despite the lethal nature of the herbicide that is being applied at the moment, is it possible that weeds could naturally develop a resistance to it over time. What would that period be likely to be?
Yes, it is possible that they would be likely to develop a resistance naturally. There is resistance to a small degree in most weed populations—even one in a million individuals—that would be selected for over time. Many weed species are now resistant to many of the conventional herbicides. The way to get round that is either to use some other technique—an alternative to chemical control—or to vary the chemicals so that the weeds are not always hit by the same chemical.
So, as this herbicide is sold as part of a package with the genetically modified crop, in 10 years' time that crop would have to be genetically modified again?
Herbicides have evolved over a century, as have crop varieties. The period for achieving resistance is 10 years, as an order of magnitude—not 100 years or one year.
I am not asking you to tie yourself to saying 10 years. However, over time the relationship between the herbicide and the crop will wear out.
It will. Yes.
It will be necessary to re-engineer the crop.
There are always developments, and it is necessary to keep ahead of pest resistance by breeding new cultivars—whether conventionally or otherwise—and using different herbicides. The way to extend the effectiveness is to trick the pest organism—whether it is a weed or an insect—by killing it or regulating its balance in different ways. GM herbicide tolerance should be no different. A sensible management of that package would mean not deploying it all the time and using other means of control as well. However, factually what you are saying is correct.
That is quite correct. Perhaps I should mention at this point that there is herbicide tolerance that has nothing to do with GM crops. There are conventionally bred, herbicide-tolerant crops to which the considerations would apply.
When, and from whom, did you find out about the contaminated Advanta crop?
I admit freely that we bought some hyola oil-seed rape seeds for trial purposes. Even though we were deeply involved, the first that we knew about the issue was from the radio. I thought, "Oops, I've got some." We tested our seeds immediately.
Dr Squire, I note from your biography that you play a prominent role in the Government's farm-scale evaluations of GM crops. You are a member of the co-ordinating group of the scientific consortium that is carrying out the trials and project manager for the spring oil-seed rape contract. I also note that Dr Ramsay has, for the past three years, supervised a project that is funded by the Ministry of Agriculture, Fisheries and Food and is aimed at understanding and attempting to quantify gene flow in oil-seed rape in the environment. Given all that, are you saying that you heard the news over the radio, not from the Executive or the UK Government?
That is correct. My involvement in the farm-scale trials does not concern the Advanta seed. I forget the precise details. The Scottish Executive tends to keep us well informed and we have good links with the Scottish Agricultural Science Agency at East Craigs. However, we were not given advance knowledge; we were just the same as any other farm that had bought the seed.
How long after you heard about the contamination on the radio did you receive a formal communication from a member of the Executive or an Executive department?
I think that we contacted Advanta first of all to ask whether our seed was contaminated; the company supplies many seed types. Advanta got back to us a day or two later to say that it was likely that our seed was contaminated.
In view of the nature of the work that you and your colleague, Dr Ramsay, have been doing, presumably a lot of information must be going back and forward to the Executive and the UK Government. Did it surprise you that you were given no prior or immediate notice when the problem came to light?
I am not sure whether it surprised us. Someone in MAFF would have had to inform us, but it is clear that no-one saw a reason to do so. I do not think that the Scottish Executive knew anything at that time. The Executive has said that it was not told that we had any seed and it would not have known that. However, within hours or days of our finding out, we were in contact with the Executive.
Did you contact the Executive or did it contact you?
I cannot remember precisely. It was a matter of hearing something and acting on it. Gavin, can you remember?
There were contacts at senior management level between the institute and the Executive. It is fair to say that we were, to an extent, in the same position as farmers. We were consumers of seed and we had seed that we had bought from Advanta, albeit for experimental purposes. We were not treated any differently.
Given that we are world leaders in the area—we have a fine reputation on gene flow—I am surprised that there was no specific contact and that our advice was not sought in advance. I was most surprised to hear the announcement.
It is not my job to protect the Scottish Executive, but in fairness I should point out that the Executive was not to know who had bought the seed.
I have two questions that have not been answered in the overall questioning. First, are you confident that the trials that are taking place in Scotland will pick up long-term impacts? Secondly, what significant level of change in the species that are measured in the trials will you be able to detect?
I will deal first with the long-term impact. Clearly, if we are commissioned to work over three years, we can detect only what happens in that time scale, then give our best estimate of how the impact will extend in the future. Clearly, our estimate cannot extend very far into the future.
Could you expand on that point?
Yes. We are working on the weeds in the field, the vegetation surrounding the field, the things that eat the weeds, and the beneficial organisms that move around the crop, including butterflies and bees.
It concerned the level of change in the species that are measured.
That is one of the most difficult things to do. The effect can be measured, but the judgment of whether it is important is subjective. We examined several major changes in husbandry that had occurred over the past century and how they had affected the populations of weeds. For example, we looked at the way in which what is called the soil seed bank—the buried weed reservoir, which is a big reservoir of diversity—decays over time. It decays and is replenished, so it is possible to detect change in that. We think change that is in the order of between 20 per cent and 40 per cent will be detectable. That is less than the extent of change that took place in the movement to autumn-sown cereal cropping or in the introduction of persistent herbicides, which was massive, or in the introduction of oil-seed rape as a widespread crop. Twenty per cent to 40 per cent might seem quite substantial over a year, but from our calculations we believe that we can estimate that.
Would it be fair to say that the investigation, although fairly widespread, is limited, as long-term effects cannot be calculated, effects that are less than 10 per cent might go unnoticed, and the sub-soil system is not being investigated at present?
That is correct. We are not yet certain that we can pick up the small effects because of variability as populations change over time by such amounts. I stress again that the effects that we are convinced we can pick up are smaller than many of the effects that have been caused by big changes in the past century.
You can do only what you have the resources and the money to do.
We have pursued the discussion as far as we want to with the witnesses, whom I thank for their patience with us. We have stuck with the witnesses throughout the evidence-giving sessions; their ability to explain some detailed matters to us in clear language has been useful.
Previous
Subordinate Legislation