House of Commons - Environmental Audit

Select Committee on Environmental Audit Minutes of Evidence

Memorandum submitted by The Friends of the Earth

Friend of the Earth inspires solutions to environmental problems, which make life better for people

Friends of the Earth is:

— the UK(s most influential national environmental campaigning organisation;

— the most extensive environmental network in the world, with almost one million supporters across five continents and over 60 national organisations worldwide;

— a unique network of campaigning local groups, working in over 200 communities throughout England, Wales and Northern Ireland; and

— dependent on individuals for over 90% of its income.

1. SUMMARY

1.1 The FSE were conceived to help the Government with a problem and keep the biotechnology industry happy. Had more time been allocated to their design then the scale of the FSE may well have been smaller in size and therefore the potential environmental impact reduced.

1.2 The statistical power of the FSE experiment meant only large differences in biodiversity could be detected. This flaw in the design could have been rectified by basing the design on what level of difference would be ecologically significant.

1.3 The direct involvement for the biotechnology companies in the management of the FSE undermined their credibility and independence.

1.4 The FSE failed to meet the Scientific Steering Board's requirement for a geographical distribution which mirrored commercial crop growing.

1.5 The failure to record yield accurately has undermined the commercial validity of the results especially because of the control the biotechnology companies had over timing of herbicide applications on the GM crops.

1.6 The impact on weeds and seed banks of herbicide used on GM oilseed rape and GM beet was dramatic and should lead to EU member states rejecting applications for Part C marketing consents.

1.7 The ban on atrazine and the introduction of glufosinate ammonium/atrazine mix to control weeds in GM herbicide tolerant maize in the USA has seriously undermined the validity of the fodder maize FSE.

1.8 The Government should integrate all its rural and farming policies and the impact that GM crops are likely to have on the whole food chain.

1.9 Key GM issues that are still not resolved despite a five year moratorium on GM approvals are coexistence, liability for economic and environmental harm, traceability and labelling, food safety assessment and environmental impact assessment methodologies. These need to be addressed before contemplating further commercial approvals.

2. INTRODUCTION

2.1 Friends of the Earth welcomes the Environmental Audit Committee investigation into the design and management of the Farm Scale Evaluations (FSE). This is a timely intervention in view of the fact that the Government will have to make important and far reaching policy decisions on the commercialisation of GM crops in the next 12 months, and that the FSE were highly controversial right across the UK and large amounts of public money were involved. The FSE results have to be placed in context. They are one piece of a complex jigsaw which the Government, along with other EU Member States, has to assemble before making the final decisions on whether herbicide tolerant crops should be licensed for sale and growing in the UK. The Government has persistently said that decisions about GM crops and food must be based on "sound science". Friends of the Earth supports this approach when dealing with parts of the decision making jigsaw that must be informed by scientific information. However, the majority of issues that complete the jigsaw are not scientific at all but economic, social, cultural, ethical and political. This is recognised by the public and came through very strongly in the report on GM nation? The Public debate[1]

2.2 There are many important issues which the FSE did not address including environmental impacts such as gene transfer in the soil, cross pollination in crops resulting GM volunteers, out-crossing to wild relatives and the fate of herbicides and their breakdown products. Many of these issues have economic, social and ethical implications as well.

2.3 Friends of the Earth considers that the motivation behind the FSE was primarily political. They were conceived in haste after the public raised objectives to the importation of GM soya and maize from the USA and inadequacies in the marketing consent risk assessment under the EU directive 90/220 were exposed. At the same time the biotechnology companies had several GM crops which had been granted EU marketing consents or were very close to achieving this. These included T25 herbicide tolerant maize (produced by AgrEvo, now Bayer CropScience) and two herbicide tolerant oilseed rapes. The latter were Plant Genetics Systems crops (now Bayer CropScience) and were only prevented from receiving marketing consents by the late decision by the French Competent Authority in late 1997 not to sign the final approval because of concerns about the adequacy of the risk assessment.

2.4 Thus the Government were under pressure from all sides and the FSE provided a convenient way out which they hoped would dissipate public concern and keep the biotechnology companies happy. Friends of the Earth's view is that the FSE should not have taken place until issues more fundamental to GM crops had been resolved if possible. These include methodologies for food safety assessment, improved risk assessment, coexistence and liability, traceability and labelling, corporate control and the lack of research into alternative solutions to the problem which GM crops are said to solve. It seemed to us that without answers to these problems the public may be very reluctant to accept GM crops. Most of these problems remain unanswered and the public remains deeply sceptical about the value of GM crops and the motives behind them. We will expand on these points as we deal with the specific questions raised in the Committee's press release of 18 September.

3. THE ADEQUACY OF THE DESIGN OF THE FARM SCALE EVALUATIONS AND THEIR ABILITY TO ANSWER THE QUESTIONS POSED AT THE OUTSET OF THE TRIALS

Concerns about GMHT crops prior to the Farm Scale Evaluations

3.1 Concerns about the environmental impacts of introducing genetically modified herbicide tolerant (GMHT) crops have come from a wide range of non-governmental organisations and scientists. Initially they focused on issues such as the increased use of herbicides, the spread of herbicide tolerance to related weeds, the development of multiple herbicide tolerance in volunteers and damage to habitats adjacent to crops. These issues have yet to be resolved and are not answered by the Farm Scale Evaluations.

3.2 The statutory conservation agencies (English Nature, Joint Nature Conservation Council, Scottish Natural Heritage and the Countryside Council for Wales) were the first Government agencies to raise concerns about biodiversity, stating at the end of 1997 that "There is every possibility that the use of GMHT crops would also lead to increased herbicide use overall and would certainly lead to "weed-free" cultivation systems over large areas of arable cropping. Coupled with a change to broad spectrum herbicides, this would have a severely detrimental effect on wild plants, insect and bird populations using farm fields and their margins."

3.3 The conservation agencies recommended a moratorium on the growing of GM crops in the UK for a period of five years. Through the Five Year Freeze alliance, this was eventually supported by over 120 national organisations, representing the full range of civil society, 50 local authorities, and many food companies. The range of concerns voiced by civil society at this time included concerns about the health impacts, ethical concerns, economic impacts, consolidation of the food chain and farming, as well as the direct and indirect impacts on the environment. It was against this background of growing public concern that the Government announced the Farm Scale Evaluations. Biodiversity impacts appear to have become the focus of examination, perhaps because they fitted within the regulatory context while many other public concerns, such as corporate control of the food chain or ethical issues, did not.

3.4 Friends of the Earth considers that the trials should not have been paid for by the public purse. In 1998, the European Commission published a proposal for amendments to the GMO Deliberate Release Directive 90/220, as a result of an emerging consensus within the EU that the safety provisions it contained were inadequate. In December 1998, at the EU Council of Environment Ministers, it was agreed that procedures to make the risk assessment of GM crops more rigorous should be adopted immediately, and in advance of any new legislation. The new procedures, which have since been incorporated into the new Deliberate Release Directive 2001/18, required a consideration of the indirect as well as the direct effects of GM crops, such as those caused by changes in crop management. In normal procedures for applicants, such research would be paid for by the biotech companies seeking to market the GM crop. As the FSE examine exactly the type of indirect effect required to be examined by applicants for marketing consent, we cannot understand why Monsanto and Bayer have not been asked to make any financial contribution towards this research.

The quality of design

3.5 The political motivation behind the FSE, and its influence on the quality of the design, is clear from the wording of the initial announcement, which stated that the Government had "reached agreement in principle with the plant breeding industry for a programme of managed development of herbicide-tolerant GM crops" (our emphasis) in which "farm scale plantings" would be "monitored for ecological effects along with comparable plantings of conventional crops." The essentials of the research were announced by a politician and the detailed specifications for the project had to be drawn up later. This was done by a process of informal consultation by the then Department of the Environment, Transport and the Regions. There is no public record of this internal consultation. It was at this stage that the objectives, scope and parameters of the research that would accompany the industry's managed development of GM crops were developed.

3.6 There is no question of the undoubted expertise of the research consortium of the ITE, IACR and SCRI who undertook the research. But the main design parameters were laid down prior to their involvement, including the period of the research, the maximum number of sites, the objectives and the null hypothesis. In fact, the contractors were only to "design and implement a monitoring programme" within the existing experimental design. The contractors have been working within an experiment that was politically motivated and designed.

3.7 The number of trials was determined by an analysis of statistical power, which is the ability of an experiment to detect difference with any certainty. As a rule, the more replications are used in an experiment, the more sensitive it becomes at detecting small differences. The sensitivity of the FSE was set at 50%, in other words they were designed so that they could detect differences of 50% between the GM and non GM crops with a degree of certainty, but smaller differences would be harder to detect, and so even if they occurred they could not be said to be statistically significant. We have always questioned the use of this fairly arbitrary level of difference. It would have been much more sensible, from a scientific point of view, to establish what level of change was meaningful in ecological terms (as opposed to purely statistical terms) and design the experiment in such a way that it was sensitive to that level of change. Such an approach would probably have required a far greater level of replication, which was not really possible within the constraints of the farm scale trials.

3.8 The impression has been given that the trials were an inevitable consequence of a smooth procedure of environmental evaluation. This is simply not true. For example, one of the results papers states, "these GMHT crops were included in the FSEs because they had already been assessed as safe in terms of human health and direct environmental impacts"[2]In fact, neither GM oilseed rape, sugar or fodder beet has yet been approved under EU procedures for assessing human health and environmental impact (as set out in Directive 2001/18 and Regulation 1829/2003). The GM fodder maize used in the trials has never been assessed for its safety as an animal feed because it was approved before this was a legal requirement. Friends of the Earth has found serious procedural irregularities in its approval under Directive 90/220 (Friends of the Earth evidence to the Chardon LL hearings).

The scale of the trials

3.9 Friends of the Earth has always maintained that the sheer scale of these trials (and the resulting enormous cost) was a result of pressure from the biotech companies for the Government to make steps towards commercial production, rather than the needs of experimental design. In fact, there is no public record that shows that the research could not have been carried out on smaller plots. The only real justification we have seen is that, in the case of the GMHT crops examined in the FSE, the "principle evidence on which to assess their ecological impact had been gained in small-field plots, from which there was little consistent indication of positive or negative effects on plants and animals of farmland"[3]But this is entirely incorrect. In fact, out of 155 applications for deliberate release of GM crops prior to September 1998, only 4 specifically mentioned environmental research as the purpose of behind the release, and these did not examine biodiversity. The rest of the small trials were agronomic and development trials, largely conducted by the biotech companies. Neither have we seen evidence (for example in marketing consent applications) that Monsanto and Bayer ever conducted ecological research on their crops.

3.10 We do not feel that the scale of the trials has been justified by the results. The most striking results are on the abundance, biomass and reproduction of non-crop plants and the abundance of those organisms that feed directly upon them, exactly the organisms that could have been studied in small plots. However, the wide-ranging invertebrates (such as ground beetles), which are commonly considered to be difficult to examine in small trials, have provided largely inconclusive results even from the scale used in the Farm Scale Evaluations.

3.11 Friends of the Earth has no doubt that the consultants who were charged with carrying out the fieldwork and analysis of the data collected in the FSE are of the highest integrity and have been scrupulous in their work within the constraints imposed upon them. However, the FSE were conceived for political rather than scientific reasons and they have had to work within this.

The ability to answer the questions posed

3.12 The importance of oilseed rape and beet crops, in particular, for wildlife in the arable rotation and hence the importance of maintaining biodiversity levels in these crops, has been shown clearly in the results of the FSE[4]

3.13 Friends of the Earth considers that the results of the Farm Scale Evaluations of GMHT spring oilseed rape and the GMHT beet crops provide enough information to show that the growing of these crops will be damaging to UK wildlife. Combined with the results of the recently published DEFRA research study entitled "Modelling the effects on farmland food webs of herbicide and insecticide management in the agricultural ecosystem (EPG 1/5/188)" there is good evidence that the impact of loss of weed seeds as a food source over the winter would cause a catastrophic decline in the numbers of already declining bird species.

3.14 Unfortunately, in the case of GMHT maize, we consider that the experiment will be unable to answer the questions posed at the start for the following reasons:

3.15 The herbicide used in the conventional halves (atrazine) has been banned because of its toxicity. It was used so much in the trials that it is doubtful that the effects of other conventional herbicides could be examined separately. This should, however, be attempted.

3.16 The yield of maize was not measured and we do not believe that the management regime in the GM halves was yield maximising.

3.17 Bayer/Aventis were not forthcoming about the commercial herbicide regime used in the United States (glufosinate/atrazine mixtures) and so the FSE did not reflect commercial reality.

4. THE CONDUCT AND OPERATION OF THE TRIALS

Yield

4.1 Friends of the Earth has always maintained that it was a serious flaw in the FSE that they did not routinely measure yield. The rationale given for this is that "these evaluations were not intended to compare the performances of the crops but rather the effects on biodiversity of the management of the crops"[5]However, the herbicide strategies used by farmers in normal practice are entirely based on the need to get a good economic performance from their crop—weed control is a balance between the cost of the herbicide, the effectiveness of the weed control it provides and the final yield. It is an extremely rare farmer who manages their crop purely for biodiversity. While the researchers state that farmers were asked to use "cost-effective" weed control, it is unclear how this could be established. Farmers in the trials were paid for taking part, not for the resulting crops, which were ploughed in or land-filled and so, not only were farmers acting in a system entirely unlike that of normal production, but they would have had no means of judging whether or not the final yield would justify the herbicide regime they used.

4.2 In addition, Bayer (then Aventis) took very rigid control of herbicide applications in GM oilseed rape and GM maize; farmers had to make herbicide decisions in conjunction with a representative of the company. In contrast, Monsanto gave training and then left the final decisions to the farmers although in one case (see later) this appears not to have been the case. The degree of control exerted by Aventis/Bayer is shown by the fact that the researchers noted that inputs to the GM beet crops were applied "more flexibly" than to the GM oilseed rape and GM maize[6]

4.3 Evidence from the United States indicates that a single application of glufosinate ammonium provides inadequate weed control in maize. Maize is extremely sensitive to weed competition in the early stages of its development[7]one of the reasons why, no matter what strategy is used, maize fields are always kept remarkably clean. In the US, agricultural advisers now recommend two applications of glufosinate and/or the use of the conventional herbicide atrazine (used on the non-GM halves of the FSE). In fact, Aventis/Bayer specifically markets a glufosinate-atrazine mixture in the US (trade name Liberty ATZ) for its GM maize. If such a mixture had been used on the GM half of the trials, then it is certain that the biodiversity of the GM maize would have been much lower than was recorded. Despite the fact that it had been marketing Liberty ATZ in the US for several years, Aventis/Bayer only informed DEFRA that this was commercial practice after Friends of the Earth raised the matter at the Chardon LL National List hearings. We have anecdotal evidence that the GM maize performed extremely poorly under the regime used by Aventis/Bayer in the FSE and we are convinced that, had yield been measured, this would have been apparent.

4.4 The researchers conducted an "internal audit" to examine whether farmers had applied the GMHT herbicides in line with the simulated label recommendations supplied by Bayer and Monsanto. However, Aventis/Bayer did not provide full information about its experience of commercial practice elsewhere, and so the label recommendations may not have reflected real herbicide regimes. The researchers state that they measured surrogates for yield, including height and development stage, and that there was little difference between the conventional and GM maize. They used this as proof that the maize trials had not been manipulated to enhance biodiversity. However, height of maize crops is a very poor indicator of yield.

4.5 All agricultural plant varieties, including forage maize, must be tested for their "value for cultivation and use" before they can be added to the UK's National List of agricultural plant varieties. These trials are managed and assessed by the Plant Varieties and Rights Office of DEFRA. The characteristics measured in these trials have been set to establish the agronomic value, in particular the yield, of the new variety. Characteristics considered to be relevant are dry matter yield, dry matter content at harvest, early vigour, cob ripeness at harvest and lodging resistance[8]None of the characteristics measured in the Farm Scale Evaluations (height, development stage) are considered important, and are certainly not used as a substitute for yield. There is no question that the measurements made in the FSE provide little useful information about the yield or agronomic performance of the GM maize. While Bayer's GM maize variety Chardon LL has been assessed through the national list procedure, the national list trials make no distinction for GMHT crops and it was grown using conventional herbicides (ie atrazine). Thus, the yield of the GM maize when grown with glufosinate has not been assessed, either in the Farm Scale Evaluations or in the National List trials.

Corporate influence

4.6 One of the main concerns in the conduct and operation of the FSE was that the companies who stood to gain most financially from a positive result in favour of GM crops were very closely involved with the selection of farms and the application of the GM herbicide. In the FSE paper by Champion et al[9]it makes it clear that "contracts were signed between the farmer and SCIMAC to grow crops as sires for the FSE". However, it is Friends of the Earth's understanding that Monsanto and Aventis CropScience (latterly Bayer CropScience) contracted the farmers who took part. Thus far from being "volunteer farmers", as erroneously suggested in the Champion paper, the FSE farmers were under contract to biotechnology companies. The fact that the companies advised on the timing of the application of the herbicide to the GM crops undermined the credibility of the FSE. Delay in applying the GM herbicide would increase the amount of biodiversity in the early part of the season. One FSE beet farmer told a fellow farmer that Monsanto had told him not to control the weeds on the GM crop despite his stated view that weed competition merited it.

4.7 Friends of the Earth believe that the direct involvement of SCIMAC, Monsanto and Bayer CropScience in the FSE tarnished the project from the start. This could have been avoided by limiting the involvement of the biotechnology companies to just supplying the seed and chemicals, applying for the part B release consents (and paying the fees to DETR/DEFRA) and funding the whole project by sending a cheque to DETR/DEFRA. Had this level of independence been established at the start of the FSE perhaps the level of public suspicion would have been less. It did not take place which further strengthens our view that the FSE were politically motivated in the first place.

Were the FSE representative of current farming?

4.8 The validity of the FSE has been furthered undermined by the lack of interest from farmers (hence the need to recruit inexperienced farmers and research farms—see below). The Scientific Steering Committee placed heavy emphasis on the requirement for obtaining farms in the FSE that ensured "representative ranges of size, intensity of management, geography and so on"[10]In the second interim report[11]they stated "farm intensity and geographic location were among the most important factors in selecting farms".

4.9 Until the raw data on the FSE is available, it will be impossible to judge how well the FSE represented the range of farm intensities. However, the goal of obtaining a representative geographic spread was not attained. For example, for spring oilseed rape, SE England was underrepresented (5% of sites against 16% of commercial growing) and eastern region over represented (35% of FSE compared with 19% of the commercial area). Recruitment for fodder maize was even less successful in terms of geographical distribution. Eastern Region hosted 38% the FSE compared with 8% of the national crop. SW region was seriously under represented in the FSE (only 12% compared with 45% of the maize grown nationally). In addition, there were several hotspots where single farmers or neighbouring farmers hosted many FSE, for example, 5 in less than 6 square kilometres in 2001 and 2002 at Hinton Waldrist in Oxfordshire and 7 in one parish in Dorset. Despite this lack of success to reach the Scientific Steering Committee's goals on geographical spread, the FSE papers state "It generally reflected very well the areas that grew these crops nationally"[12]They go on to state "for forage maize, although somewhat fewer sites were offered in the Southwest than typical, the achieved replication of 12 sites, together with 16 in the Midlands and West, was adequate to detect any interaction between treatment and environment should any have occurred. Similarly, the small number of sites for spring oilseed rape in the Southeast was offset by the larger numbers in the neighbouring Eastern region".

4.10 It is clear the Scientific Steering Committee and the consultants had a very different view of the importance of geographical spread of the FSE. In the case of fodder maize, no FSE were grown in Cornwall, Devon and Somerset which are the three top counties for growing the crop.

4.11 The FSE crop management and agronomics were reported in the paper by Champion et al[13]This detailed how the sites were selected and the criteria included "the variation in intensity of farm management" and "usual agronomy". From this we infer that "usual agronomy" meant the methods previously used to grow the crop under trial on that farm. Farmers were asked to complete a self assessment form on cropping intensity. This suggests that farmers taking part in the FSE should have grown the crop that they were trialling before. Friends of the Earth have established that this was not the case in all the FSE. The fodder beet FSE at Blaisden in Gloucestershire in 2001 took place immediately after the farm (a stud) had changed hands and the field used was previously under permanent grass. Two trials took place on the same field at Witham-on-the-Hill Lincolnshire—spring oilseed rape in 2002 and fodder maize in 2002. Local residents reported to Friends of the Earth that the field had not been used for oilseed rape for a least a decade and had been in permanent set aside. In addition, the field was used for tipping potato washings and other organic wastes. The owner has never grown fodder maize before and does not own any livestock. It is fair to say then that this FSE does not represent the commercial crop farming as is normally practiced and it would be hard to imagine what information was included on the self-assessment form. The land has now returned to set aside. One FSE farmer in Shropshire told Friends of the Earth's Senior Campaigner that despite hosting an oilseed rape FSE, he had never grown the crop before. These examples illustrate that the FSE were not necessarily on farms that were experienced in growing the crops in question and previous management of the crops would not have been available. Many FSE also took place on experimental farms where the cropping pattern and management could have been different than those followed in a normal commercial farm, for example Advanta Seeds UK trial grounds at Boothby Graffoe (Lincs), ADAS sites at Itchen ADAS (Hants), ADAS Meden Vale (Notts), ADAS Preston Wynne (Hereford) and IACR (Herts).

4.12 The results of the FSE for fodder maize have been further undermined by the EU announcement that atrazine, the herbicide used on 75% of the non-GM crops in the FSE, is to be banned under Directive 91/414. It is unclear whether the consultants or Scientific Steering Committee were aware that atrazine could be banned. The statistical validity of using the remaining sites was not covered in the FSE report nor the possibility that the ban would result in new weed control methods being used on fodder maize which may have a different impact on biodiversity than atrazine or other herbicides currently approved for maize.

4.13 Additional concerns about the validity of the maize FSE arise from doubts about the efficacy of glufosinate ammonium, the herbicide used on the GM fodder maize. These doubts arise because of the decision of Aventis/Bayer to introduce a glufosinate ammonium/atrazine mix (Liberty ATZ) in the US maize belt because glufosinate ammonium alone was not providing farmers with the weed control that they were expecting from GM herbicide tolerant maize. It is unclear from the FSE reports when the Scientific Steering Committee and consultants became aware of this change in the USA. It first emerged in the UK at the Chardon LL hearings[14]and on BBC 2's Newsnight (25 June 2002). There is no evidence that Bayer CropScience informed the consultants or the Scientific Steering Committee of the development of Liberty ATZ in advance of this. Once again the validity of the maize trials is brought into question. At present there is no data in the public domain relating to the efficacy of glufosinate ammonium used in the UK.

4.14 One way in which the Environmental Audit Committee could clarify issues relating to the conduct of the FSE would be to ask the Scientific Steering Committee chair, Professor Chris Pollack to give oral evidence.

5. THE WAYS IN WHICH THE RESULTS WILL BE INTEGRATED WITH POLICY AND DECISION MAKING

European Approvals

5.1 Friends of the Earth considers that the results of the FSE for GMHT oilseed rape and the beet crops show that these GM crops are harmful to biodiversity and should not be grown commercially. The consideration of applications for marketing consent for these crops is ongoing and these applications should be rejected. The Deliberate Release Directive 2001/18 requires that member states act in accordance with the precautionary principle with respect to GMOs (Art 4.1) and that the assessment of environmental risk must consider "adverse effects [that] may occur directly or indirectly through mechanisms which may include . . . changes in management, including, where applicable, in agricultural practice" (Annex II). At present, the marketing applications are still being considered by the rapporteur member states; Belgium, Germany and Denmark. The UK government should forward the results of the FSE to these member states and should recommend that the applications be rejected.

5.2 It is quite clear that the harm caused to the environment by the GM linked herbicides cannot be extracted from the GM crops. The marketing of the crops as being "Liberty Link" and "Roundup Ready" makes this link to the herbicides quite clear. The fact that three sets of results were observed when only two herbicides were used shows that this is a crop-herbicide interaction, not just a result of the herbicide alone. The researchers themselves note this, stating that "Many species showed similar responses to the GMHT treatments in beet and spring oilseed rape, where two different herbicides were used. This contrasted with their response in maize, where the same herbicide as in spring oilseed rape was applied. This suggests that in this experiment it was the interaction between the species and the management associated with a particular GMHT crop rather than sensitivity to the specific herbicides used that determined species responses"[15] (emphasis added).

5.3 Put simply, under the terms of the Deliberate Release Directive 2001/18 GMHT oilseed rape and beet crops should not be approved. In the case of maize, the results of the FSE are inconclusive, but as we have discussed, we consider this to be an artefact, rather than a real finding.

5.4 It is now being suggested that changes in wider farmland management could offset these effects, particularly an increase in winter stubbles. Such pie-in-the-sky suggestions should not be a consideration in the legal decision-making process. Firstly, we cannot envisage how such changes could be enforced or implemented in practice:

— How could the biotech companies possibly guarantee that farmers who bought their products left winter stubbles?

— Would this be a condition of sale, and if so how could it be enforced?

— What proportion of their land would have to be left to winter stubble to offset the harmful effects of the GM crop?

5.5 While Friends of the Earth is in favour of a general switch in cropping from winter to spring sown crops, we believe that this should be done to improve biodiversity overall, rather than to offset the harmful effects of introducing GMHT crops into the arable rotation.

5.6 Secondly, there is evidence that the harmful effects of GMHT crops can't be offset in this way. While the results of the FSE are narrowly focused, the DEFRA modelling study[16]did look at other management changes and concluded that increasing winter stubbles has no effect on the important bird food-plant Chenopodium album (fat hen), as it cannot produce seed over the winter. When both GMHT rape and beet crops are introduced, all weeds declined, with fat hen approaching extinction within fifty years. "The consequences of introducing GMHT sugar beet were extremely severe, with a rapid decline, and extinction of the skylark within 20 years. " Increasing the area of winter stubbles made no difference to this decline.

5.7 Similarly, it has been suggested that herbicide tolerant crops could be managed to enhance wildlife by delaying weedkiller application[17]but the results of the FSE show that this would not increase winter seed production, even in maize, and in the GM beet and oilseed rape, seed production was severely reduced. Survival of chicks reared in the summer will depend on over winter food supplies, as will the breeding fitness of adults the next year. A recent study found that in winter a narrow range of weed seeds made up the entire diet of birds such as the skylark, dunnock, greenfinch, linnet and reed bunting[18]

5.8 The results of the Farm Scale Evaluations address only one of the many questions and concerns that have been raised with respect to GM crops: for example, they do not address the safety of these GM crops for food and feed use; they do not address the likely consequences of genetic pollution from these GM crops and the impact on receiving populations of wild plants; they do not address issues around co-existence and potential gene stacking; they do not address the environmental consequences of the increased use of glufosinate ammonium and glyphosate.

6. THE IMPLICATIONS OF THE TRIAL RESULTS FOR GOVERNMENT AND OTHER DECISION-MAKERS

6.1 One of the key aspects of the risk assessment required under the GMO Deliberate Release Directive 2001/18/EC, Annex II of the Directive (Section D2) requires that the applicants provide evidence on the following potential environmental impacts of the GMO.

Possible immediate and/or delayed effects on biogeochemical processes resulting from potential direct and indirect interactions of the GMO and target and non-target organisms in the vicinity of the GMO release(s).

Possible immediate and/or delayed, direct and indirect environmental impacts of the specific cultivation, management and harvesting techniques used for the GMHP where these are different from those used for non-GMHPs.

6.2 These are precisely the effects that the FSE were designed to look at and any crop which shows any indirect effect of this type should automatically be refused a Part C marketing consent. It is also a good reason for the biotechnology companies to have paid for the FSE not the tax payer.

6.3 Under the previous Directive 90/220, requirements under the risk assessment did not include indirect impacts. It is worth remembering that two varieties of GM oilseed rape were merely awaiting the signature of the French Government (as competent authority for those applications) before they were granted Part C consents. It was the timely intervention of the Joint Nature Conservation Committee (led by English Nature) that halted these approvals in 1997 when they pointed out the use of glyphosate and glufosinate ammonium (both broad spectrum weedkillers) on GM herbicide tolerant crops could exacerbate the impact of conventional farming by further destroying weeds which form the base of the farmland food chain and therefore result in more serious decline of farmland birds.[19]

6.4 Thus the FSE will be of great interest to the member states dealing with the current applications for Part C consents under 2001/18/EC. The current applications are being dealt with by the following member states:

Spring Oilseed rape—Belgium.

Fodder Beet—Denmark.

Sugar Beet—Germany.

Sugar Beet—Belgium.

6.5 The fodder maize (T25) was granted a marketing consent under 90/220 Directive in 1998 by France before the de facto moratorium was agreed. This consent should be reviewed following the FSE results as consent holders are required to inform the competent authority of all new information.

6.6 The competent authorities are required to assess new applications to ensure that they meet all the requirements of the risk assessment. The FSE results should be submitted by the applicants as an addition to the risk assessment dossier. The Belgian authorities have delayed their decision on the application for the spring oilseed rape from Bayer CropScience so that they can consider the FSE results from the UK.

6.7 If the Part C marketing consents progress beyond the competent authorities, the UK government will have to issue an opinion on the whole of the application (including the FSE results) based on the advice of the Advisory Committee on Releases to the Environment (ACRE).

6.8 One variety of T25 fodder maize has progressed through the Seed Listing process (The National List of Varieties Regulation 1992) to the point where it is awaiting a joint decision by the four administrations of the UK before it is placed on the National List of Varieties (the Seed List). The devolved administrations in Wales and Scotland should take particular note of the FSE results for the fodder maize because yield data they have from National List Trials did not include the use of glufosinate ammonium in the manner it was applied to the GM crops in the FSE. They may well wish to look at the National List Trial data again because it does not represent how the crop would be grown commercially. The fact that there is no reliable yield data from the maize FSE will make decision making on seed listing all the more difficult.

7. THE COSTS AND BENEFITS OF GM FOOD, BEARING IN MIND THE POTENTIAL MARKET, IN THE LIGHT OF THE FARM SCALE TRIALS AND THE RECENT STRATEGY UNIT REPORT

7.1 Friends of the Earth agrees with the conclusion of the Prime Minister's strategy unit report on the costs and benefits of GM crops[20]Paragraph 4.4.7

In the short term, negative consumer attitudes can be expected to limit the demand for the products containing GM foods, and therefore the economic value of the current generation of crops.

7.2 This conclusion is supported by the current position of the major supermarkets[21]British Sugar[22]and the recent announcement by the Co-op to continue their prohibition on GM ingredients in their own brand food, growing GM on their farmland and investments in GM in agriculture[23]

7.3 During the last 12 months, 29 local authorities across England have joined the National Assembly of Wales in seeking to keep their area free of GM crops and foods by excluding GM food from their purchasing and GM crops from being grown on land they own. In addition some councils have agreed to make representations to the European Commission to seek conditions on Part C consents to exclude their land area from the area where the GM crop may be grown. Such applications would need to be backed by data showing why such an exemption would be required to protect the local environment or economy[24]

7.4 The FSE results alongside the Strategy Unit's conclusion on the lack of market for GM crops in the near future, the GM science review's acknowledgement of scientific uncertainty and gaps in our knowledge and public opinion buys the Government time. This time should be used to ensure that the Government's policies on rural affairs are fully integrated and joined up. This does not appear to be the case at present. On the one hand, the Prime Minister, backed by Lord Sainsbury, appears more interested in the economic potential of GM crops than the potential economic, environmental and health impacts. On the other hand, there are Defra ministers who are intent trying to roll out the proposals contained in the Report by the Policy Commission on the Future of Farming and Food[25]and the Organic Action Plan[26]In both of these there is a recognition that farmers need to listen to consumers and respond to their demands if the rural economy is to recover and be able to operate at a profit and farm in a sustainable way. At some stage the Government's GM policy and wider rural and farming polices have to be integrated. This means that the incompatibility of GM and other forms of farming on a small island has to be acknowledged.

7.5 In this context it is worth reflecting on the current value of the agricultural biotechnology industry in the UK. This is often confused by the Prime Minister and other Ministers with the total value of all the biotechnology activity, including pharmaceuticals and medical applications. The true value of the agriculture biotechnology sector was given in a recent Written Answer to Roger Williams MP. In 2000 it was £160 million employing 1,147 people and in 2001 £198 million employing 1,145[27]This needs to be compared to the economic value of pollination services provided by beekeepers which was last estimated in 1996 to be £202 million[28] (excluding honey sales). This puts the value of the agricultural biotechnology industry into its true context by comparing it with an industry whose product integrity will be directly undermined by the commercialisation of GM crops. During the FSE, the British Beekeepers Association recommended that beekeepers move their hives six miles from the nearest GM crop to avoid the presence of GM pollen in their product.

November 2003