GM sugar beet study deception

Whilst there are many brilliant and dedicated minds in the scientific world there are increasing concerns that their work is becoming easy prey for less than honest exploitation by commercial interests - particularly when it comes to presenting industrial agendas through the media.

Although not yet picked up by the popular media, full details of the extent to which the work of scientists can become subject to blatant presentational manipulation by industry have recently come to light in a case in the biotechnology sector. This follows the publication of a study on transgenic sugar beet in the journal 'Pest Management Science'.

Although this paper was only officially released in April 2000, Monsanto had been using parts of this work prior to its peer-review and formal publication to make misleading statements about the environmental 'benefits' of its transgenic sugar beet since 1998. Further details of Monsanto's manipulation of the findings of this work, subsequently unearthed by New Scientist, are given at http://www.btinternet.com/~nlpwessex/Documents/monsanto_sustainabilty.htm : "The biotech industry is developing two very different sales pitches for its products - one for farmers and one for the rest of us," NEW SCIENTIST, Oct 1998.

A few months earlier large sections of the farming and general media had been taken in by this deception: "Genetically engineered crops can save farmers money, reduce chemical spraying and create a better habitat for birds and insects, scientists claimed yesterday", reported an excited LONDON TIMES 25 August 1998 under the title "Modified crops 'help man and wildlife'" - even though we now know that the 'findings' concerned had been neither completed nor published at that date.

Now, however, full details of the work used to make these claims are - nearly two years after the first press reports - at last in the public domain. They show that the delayed herbicide application approaches (which were high-profiled by Monsanto at the time as providing valuable environmental 'benefits' related to insect populations), using glyphosate on the transgenic sugar beet concerned, produced heavy yield losses for farmers. These approaches are therefore most unlikely to be used in practice. In fact when such approaches were used even the claimed environmental benefits themselves were not conclusively demonstrated in the study, with the authors acknowledging that: "Further studies using more rigorous ecological sampling methods" are required to monitor the full impact on insect populations.

In addition IACR has research links with more than 40 countries throughout the world.

For an analysis of the social implications of the way this latest paper from IACR on Monsanto's transgenic sugar beet has been presented to the pubic via the media, please see para 4) of the "concluding observations" at the end of the commentary below (which begins with a more technical examination of the paper).

The degree to which the scientific world has become dominated by commercial interests is further illustrated in this case by the fact that no sanction has been taken against Monsanto for its pre-publication propaganda exploitation in August 1998 of the scientific data emerging from the study.

By contrast, the internationally respected nutrition scientist, Dr Arpad Pusztai, was immediately suspended from his job at the Rowett Research Institute in the same month for releasing preliminary data on the food safety of transgenic potatoes (work subsequently published in the Lancet). This was data which Dr Pusztai considered to be of urgent and vital public importance necessitating its early release (something which could not be said in relation to the claimed 'benefits' of Monsanto's trangenic sugar beet). The urgency was acute given the large range of superficially tested GM products whose launch onto global food markets was immanent at that time.

Meanwhile the pressures that modern scientists are now subject to from political and commercial quarters have been recently examined in a paper published in 'Research in Social Problems and Public Policy', Volume 7, (Stamford, CT: JAI Press, 1999), pp. 105-135 entitled 'Suppression of dissent in science' (Martin, B). Reviewing the political and industrial sociology of modern day science the study concludes: "Studying suppression has the potential to reveal much about the dynamics of expertise, power, and legitimacy in contemporary society, but this type of investigation is bound to remain controversial itself both because of definitional and methodological challenges and because it draws attention to an exercise of power that those exercising it would rather pass unnoticed."

How much of modern day science is therefore really about 'science' (knowledge), and how much of science is really about 'interests' (power)?

The story of this Monsanto 'funded-and-spun' transgenic sugar beet study is just one example which may give us a clue as to how the direction of modern 'science' is really determined.

"All policymakers must be vigilant to the possibility of research data being manipulated by corporate bodies and of scientific colleagues being seduced by the material charms of industry. Trust is no defence against an aggressively deceptive corporate sector,"
THE LANCET, April 2000
http://www.netlink.de/gen/Zeitung/2000/000409.html

Commentary by Mark Griffiths BSc FRICS FAAV
on the paper
"Delayed control of weeds in glyphosate-tolerant sugar beet and the consequences on aphid infestation and yield"
authored by
Alan M Dewar, Lisa A Haylock, Kathy M Bean, Mike J May IACR-Broom's Barn, Higham, Bury St Edmunds, Suffolk IP28 6NP, UK
and published in
Pest Management Science, Vol 56, Issue 4, 2000. p 345-350 (April 2000)
(abstract at http://www3.interscience.wiley.com/cgi-bin/abstract/71000396/START )

This commentary provides initial observations on the above paper. The observations are provided neither by a qualified sugar beet agronomist nor by a qualified entomologist. They should therefore be seen as a basis for exploratory on-going discussion with the paper's authors, and not necessarily as definitive statements about the significance of this study.

1) The study looks at five weed management (including "no treatment") scenarios for a cultivar of Monsanto's genetically modified Roundup Ready (glyphosate resistant) sugar beet.

2) The study looks at such scenarios for the beet grown on a relatively rare soil-type prone to heavy weed infestation.

3) The study does not involve any other sugar beet varieties - i.e. there are no non-GM controls and therefore the study does not provide a comparison between GM and non-GM crop performance. It purely looks at the relative merit of different weed management strategies for the husbandry of this particular GM cultivar of sugar beet on a particular soil type.

4) Whilst of general interest, the study of itself does not provide explicit information as to whether it is economically beneficial for farmers to grow such a variety in these conditions because:

5) In addition to looking at other effects such as crop yields the study attempts to examine the impact of the various weed management regimes on populations of a variety of insects, chiefly aphids. In-depth interpretation of the ecological significance of these impacts does, however, seem to be beyond the scope of the study and clearly requires further investigation.

6) It appears that no aphicides were applied during the course of the study. In practice there are likely to be many farmers who do not wish to adopt such a non-interventionist approach, particularly where in-crop colonisation levels exceed those for chemical control measures recommended by MAFF. In such circumstances the influence of the different weed management regimes (including the use of glyphosate resistant sugar beet) on aphid numbers will be largely irrelevant. (Separate informal discussions with 'LEAF' [5] would seem to suggest that uptake of natural methods of aphid control using integrated crop management techniques is not widespread amongst sugar beet growers.)

7) The agronomic significance of crop aphids is that in addition to feeding on crops they may also transmit plant viruses which can have a significant effect on crop performance.

8) This paper makes it clear that at least two applications of glyphosate are required with the use of such GM crops - some of the study data, however, would seem to imply that in practice three applications may be required for profit-maximising farmers (although no trials were conducted on this basis, such a three-applications scenario had been previously acknowledged as a possibility by another IACR-Broom's Barn sugar beet research scientist in an article published in Farmers Weekly 20 February 1998).

9) Although carried out by scientists at the Institute of Arable Crops Research (IACR) the study was funded by Monsanto plc.

10) Below are more detailed observations on the various titled sections of the paper. In addition to 'no-treatment' and 'conventional treatment' scenarios the study examines the impact on crop yield and insect populations of three different glyphosate spray treatment regimes. All glyphosate regimes comprised two application of glyphosate as follows:

Regime Description	Beet growth stage
	first application	second application
Glyphosate 1	2-4 leaves	12-14 leaves
Glyphosate 2	8-10 leaves	12-14 leaves
Glyphosate 3	12-14 leaves	16-20 leaves

'The advent of herbicide-tolerant sugar beet....... offers the potential to reduce the damage caused by pests on young plants.'

It is not clear exactly what this statement means or how it is supported by the study data. The only glyphosate scenario which achieved comparable yields to the conventional herbicide management practice appears to have produced considerably higher aphid colonisation on sugar beet plants (whilst this might be welcomed by ecologists it would not necessarily be welcomed by farmers who might then be prompted to use an aphicide in response).

In all glyphosate usage scenarios higher weed ground cover arose (again this might be welcomed by ecologists, but not necessarily by farmers who in practice might feel prompted to apply a third application of glyphosate in response when growing the crop commercially. The later scenario was not explored in the study).

The soils in which the trial beet were grown can be particularly prone to weed infestation. They do, however, only have relevance to 9% of the UK sugar beet crop.

The control 'conventional' weed programme used in the study involves five herbicide applications (table 1), whereas the study indicates that on more normal soils only three applications would be likely to be needed.

In this study it is stated that (as measured by active ingredient) "the total herbicide burden applied in the glyphosate programmes was 45% less than in the conventional" scenario. However, if the later is adjusted to the norm of only three herbicide applications (taking out the initial application of diquat/paraquat and one of the two duplicated multi-component applications) the total amount of active ingredient under the conventional regime would fall to 8% below the level of the regimes utilising two glyphosate applications, or to 39% below the level of a three glyphosate application regime (not tested here).

However, such measurement does not take into account the relative environmental toxicity of the chemicals used. The use of glyphosate may provide an advantage here, although its ubiquitous use may also accelerate the development of weeds resistant to it and ultimately limit its usefulness as a herbicide.

The treatment regime for the conventional programme does not appear (although this is not clear) to include a residual soil acting herbicide which would provide long acting weed control. This may be because such herbicides are less suited to use on this type of high organic-matter fen soil. Nonetheless this type of herbicide may be used in other sugar beet growing situations on mineral soils.

From the farmer's point of view, however, it is nonetheless clear that the conventional regime gave the best weed control (in terms of reduced weed ground cover - table 2), especially in the early days after crop emergence. This is the time when competition between crops and weeds can be most critical and farmers may not be entirely happy with reduced weed control at this point, even when they are using the early application glyphosate 1 scenario (leaving 29.2% weed ground cover compared to 8.3% with the conventional regime). This is a factor which may encourage a third glyphosate application at an additional earlier date if and when the crop is grown commercially.

The level of aphid numbers on beet appears to be considerably higher under the glyphosate 1 regime than under the conventional regime (table 3). The glyphosate 1 regime is the one most likely to be used by farmers because of its improved yield effects (table 5) compared to other glyphosate regimes. It is possible that this high level of aphid colonisation may encourage farmers to apply an aphicide.

There is no data relating to numbers of aphid predators on the sugar beet (unlike the later weed data), so it is not apparent what effect the different herbicide regimes have on this aspect.

If increased aphid numbers (together with predators and parasites) on weeds equates with reduced aphid numbers on beet then that may be an advantage as far as the farmer is concerned. However, the implications of this data are difficult to interpret because of the statistical methods used. Nonetheless, from an agronomic perspective it is in any case difficult to see the usefulness of this data as it is the number of aphids on the beet which are critical to the farmer. In the case of glyphosate 1 treatments aphid numbers on beet appear to be higher (table 3) than when using a conventional herbicide programme.

At the same time the superior (again from the farmer's point of view) weed control obtained from the conventional regime is indicated by the fact that it is the only case where "no samples [were] taken due to lack of weeds" (table 4).

No data concerning aphid numbers on weeds relating to glyphosate regime 3 is provided.

The use of glyphosate regimes does not seem to have produced an increase in aphid predators on weeds compared with the control regime where weeds had no herbicide treatment at all - rather they seem to have resulted in a reduction in most cases. Most of the data seems to indicate zero or close to zero presence of predators on glyphosate treated weeds.

It would be interesting to see how this would compare with a regime which deployed integrated pest management techniques in sugar beet.

It is noted that predator eggs rather than predators themselves were recorded. Is it possible that any of the herbicide regimes could negatively affect egg hatch success rates?

The study concludes that: "No significant differences could be detected in predator or parasite numbers between the sampled plots on either date, although of course there would have been few, if any, in the conventional plots due to lack of weeds." Unfortunately, however, no data is given in relation to aphid predators or aphid parasites on the actual beet themselves.

Some figures for predator and parasite levels on weeds are given in table 4. This data, however, appears to be devalued by a statement at the end of the study which adds that "the sampling procedures used here were not designed to estimate populations of these latter insects quantitatively".

The general conclusions as to the influence of the glyphosate weed management regimes on these potentially beneficial insects is accordingly vague: "It was our impression ... Further studies using more rigorous ecological sampling methods are required to confirm this."

There is no data in table 4 concerning predators and parasites relating to glyphosate regime 3.

And to reiterate, there is no data relating to numbers of aphid predators and parasites on the sugar beet itself for any of the regimes (table 3).

'Virus Yellows' were not a significant problem in any of the trials. Elsewhere in the study it is acknowledged that the incidence of infection in the trial year (1998) was relatively low.

Nonetheless aphids can be important vectors for viral infection in disease active years. The raised aphid levels on beet in the glyphosate 1 treatments may therefore encourage the use of aphicides by farmers. Aphicides can have particularly damaging effects on food supplies for other animals in the ecosystem.

According to MAFF [4] : "Virus yellows, introduced and spread mainly by green aphids, is the most important disease of beet. Early infection can decrease yield by 40-50%. The disease is caused by beet yellows virus (BYV) and/or beet mild yellowing virus (BMYV)." ( http://www.britishsugar.co.uk/bsweb/growers/pests.htm )

In late spring and early summer, British Sugar advisers inspect crops daily and advise when and if to spray in their area. When plants have less than 12 leaves, one or more wingless green aphids per four plants justifies a spray; this can occur as early as the two true-leaf stage following a mild winter and early dispersal of aphids. Plants with more than 12 leaves do not need spraying until they have more than an average of one wingless aphid per plant.

Interestingly it would appear in these trials that the only weed treatment scenario to produce aphid levels on the beet above these thresholds is the glyphospate 1 regime - i.e. the herbicide tolerant regime most likely to be used by farmers planting this particular genetically modified cultivar. This would appear to indicate that the introduction of glyphosate tolerant beet varieties may lead to an increase in the use of aphicides.

Spray regimes against green and black aphids recommended by MAFF [4] include the following chemicals: Gaucho, Aztec, Evidence, Patriot, Decisquick, Dovetail and Aphox (chemical names available at http://www.britishsugar.co.uk/bsweb/growers/glossary.htm ).

Glyphosate regimes 2 and 3 reduced yields very substantially - by 24% and 31% respectively compared to beet grown under a conventional spray regime (table 5). No farmer, therefore, is likely to use such a regime. Any 'environmental benefits' to be derived from such programmes (which do not appear to be clearly specified or explained in this paper) are therefore likely to be largely irrelevant.

Users of such genetically modified sugar beet (assuming that they are willing to do so because they believe it out performs normal sugar beet - an issue also not addressed in this paper) will inevitably opt for the glyphosate 1 regime because of its considerably higher crop yield outcome (table 5) compared to other glyphosate regimes. It is even possible that they will go for yet another glyphosate programme using three applications - an option also not explored in this paper.

It should be recognised that the conventional spray regime in the study also required some hand-hoeing to achieve acceptable levels of weed control.

Nonetheless the general levels of weed control (as measured by % weed ground cover in table 2) prior to the date of first hand-hoeing (given in table 1) were superior in the conventional regime compared to the dual application glyphosate regimes.

"... the number of aphids colonising beet was reduced where weeds provided some cover to young seedlings". It is not clear how 'young seedlings' is defined and what data supports this statement (the only data on beet aphid numbers given in the study is in table 3). In fact the opposite appears to be true as far as beet aphid numbers were concerned on 19th and 29th of June (the only dates for which data is provided) in the very glyphosate regime most likely to be used by farmers in practice because of its crop yield implications (glyphosate 1).

"The effect of weeds in this study was substantial - they reduced it by 71% compared to conventional plots." It is worth pointing out that it was only the trial receiving no weed treatment at all which produced this overwhelming level of yield reduction. Nonetheless glyphosate regimes 2 and 3 produced yield reductions of 24% and 31% respectively compared to beet managed under the conventional herbicide programme. However, for some reason the authors refer to the 24% reduction as 'substantial' but 'not significant'.

1) Most, if not all, of the glyphosate scenarios explored in this study are unlikely to reflect the actual use of such herbicide tolerant sugar beet in practice once farmers have gained a few seasons experience with it. The reasons why this is likely to be the case are as follows:

2) It is difficult to reconcile the study data with parts of the study's concluding sentence which includes the statement that: "The use of glyphosate ... offers the possibilities of manipulating the habitat to the benefit of the environment ..... without loss of yield."

If it is taken that the suggested environmental benefit arising out of habitat manipulation in this study is the use of glyphosate to increase non-beet aphid numbers on weeds (thereby, so it appears to be implied, increasing by attraction aphid predators and parasites generally in the field) then the results appear to be non-definitive, and in the case of the glyphosate 2 and 3 regimes they are clearly associated with heavy decreases in yield. (It is these two latter regimes which were particularly paraded in the press prior to peer review of the study - see 4 below).

3) A number of matters in this paper are difficult to interpret. Further clarification through discussion with the authors should be encouraged.

4) Nonetheless, whatever the real implications of this study turn out to be there must be concern that widespread press reports relating to this work appeared in the farming, scientific and general press prior to the completion of its data collection, let alone prior to the submission of the paper to a scientific journal for peer review. In this case the concern is especially acute because:

1. According to MAFF weed beet are any unwanted beet within and between the rows of sown beet, in other crops, or on waste ground. They grow from groundkeepers or from seed shed by bolters. The bolters may be in sugar beet crops, may be weed beets in other crops or come from groundkeepers. As seedlings, weed beet are indistinguishable from sugar beet but many bolt easily and are prolific seed producers.

If not controlled weed beet seedlings will bolt, compete with the crop and set still more viable seed.

Sugar beet bolt in long days after experiencing cold. The plants which bolt in a root crop are the least bolting-resistant members of the variety. These can inter-pollinate and produce much seed in the English climate. Over many years this natural process has selected for easy-bolting types and these readily inter-pollinate with bolters arising either from seed contaminated by annual types or from groundkeepers.

All bolters are potential sources of weed beet if they flower early enough to produce viable seed. Although sugar-beet varieties have always given a few bolters, long rotations and hoeing probably controlled the problem until the mid 1960s. At about that time seedsmen started to make hybrid monogerm varieties using tetraploid pollinators which are less able to avoid contamination by pollen from diploid wild annual beets. Also, farmers started to close rotations (so getting fewer intervening crops in which weed beet could be controlled) and began using herbicides instead of the hoe to control weeds in their beet crops. These herbicides cannot differentiate between crop and weed beet.

Over the last 25 years these factors have allowed the balance to tip sufficiently for beet to become a sucessful weed.

Weed beet compete with crop beet and depress sugar yield. When the numbers of weed beet equal the numbers of crop beet, at best, sugar yield is halved. Even a few weed beet depress yield slightly and, if not controlled, each can produce about 1,500 viable seeds, many of which survive so that the problem becomes progressively worse with each rotation. Weed beet are woody and, if delivered to the sugar factory, create problems in slicing and sugar extraction. No single control method is likely to be successful, but a combination should be effective.

"I think we have to assume that we will get herbicide-tolerant volunteers. Their numbers are related to the amount of seed shed by the crop. In sugar beet this is low but we do get bolters. We may find we need to be controlling bolters to limit pollen spread rather than seed return and this may be much more difficult to manage." (Mike May is one of the authors of the sugar beet study discussed in the commentary above.)

"Roundup and Liberty are very efficient plant killers... Initially the new chemistry will control weed beet, but eventually [weed] beet will become tolerant to both chemicals."

4. See: " Sugar Beet: A Grower's Guide", Fifth Edition, Edited by K.W. Jaggard IACR-Broom's Barn, M. Limb and G.H. Proctor British Sugar plc, published by The Sugar Beet Research and Education Committee, Ministry of Agriculture, Fisheries and Food, London.
( http://www.britishsugar.co.uk/bsweb/growers/index.htm )

5. LEAF (Linking Environment and Farming) encourages the uptake of Intergrated Farm Management (IFM) through the development of practical guidelines and the promotion of the LEAF Audit. It promotes IFM to a broad range of groups through its national network of Demonstration Farms ( http://www.farmeco.co.uk/directory/news/showstory.asp?uniqueid=1028 ) .