Science in Society Archive

Bogus Comparison in GM Maize Trial

The research paper claiming that GM maize is better for the environment than non-GM even if atrazine is not used is highly misleading. Prof. Peter Saunders and Dr. Mae-Wan Ho report

Caught between fierce opposition from the public and heavy pressure from the biotech industry, the UK government agreed in 2000 to fund the 'Farm Scale Evaluations' (FSEs) at a cost of £3 million to the taxpayer. Three genetically modified herbicide tolerant (GMHT) crops - maize, oilseed rape and sugar beet - would be grown side by side with conventionally managed counterparts, so they could be compared.

The FSEs were severely criticised for being rigged in favour of the industry right from the start. First of all, organically managed crops were not included in the evaluations, nor were the crops grown under other low-input, integrated management regimes.

The FSEs were not intended to address safety issues, as these were assumed to have been satisfactorily resolved. There would be no evaluations on the risks of gene-flow, nor threats to a whole range of wildlife, livestock or human beings, nor effects on the soil ecosystem. There would be no data collected on yields or other important agronomic indicators.

The FSEs would estimate the effects on biodiversity using only a few indicator species of weeds and insects; and if these proved to be the same, or "substantially equivalent", then the GM crops would be given the go-ahead.

It is rather like giving a MOT certificate to a car just by checking that the tyres are OK.

But to everyone's surprise, when the results were published, it turned out that GM oilseed rape and sugar beet had a more deleterious effect on biodiversity than conventionally managed crops. GM maize, however, appeared to do better than the conventional maize crop. The Government therefore announced that it would permit GM maize to be grown commercially but not the other two.

This was a very convenient result. It allowed the Government to portray itself as being very cautious and responsive to scientific evidence and, at the same time, let the GM lobby go ahead with commercial growing of GM crops. What is at stake is the principle that GM crops can be grown commercially in the UK, and it matters little whether it is maize, oilseed rape or sugar beet. Once one GM variety has been agreed, then more will follow, if for no other reason than that pollen from GM variety will pollute the non-GM varieties, as is already happening in North America and elsewhere.

Unfortunately, there was a snag. Almost all the conventional GM maize had been treated with atrazine or other triazine herbicides, and just as the results were announced, the EU banned these herbicides on environmental grounds. This meant that the maize trial results were no longer valid.

Then, just before the environment secretary Margaret Beckett was due to give the official go-ahead for the GM maize, a paper that claimed to rescue the Government's case was rushed online in the high prestige journal Nature. It bore the confident title: "Ban on triazine herbicides likely to reduce but not negate relative benefits of GMHT maize cropping." The eleven authors of this paper come from a whole collection of Government-funded Institutes: Rothamsted Research in Harpenden, Hertfordshire, NERC Centre for Ecology and Hydrology, Lancaster, Cumbria, Broom's Barn Research Station in Bury St. Edmunds, Suffolk, NERC Centre for Ecology and Hydrology in Huntingdon, Cambridgeshire and Scottish Crop Research Institute in Dundee, Scotland.

The claim is that, according to further statistical analysis, the ban on triazine herbicides, although it might reduce the benefits of GM maize, is unlikely to cancel them out altogether. This would be a very significant result, if only it were true.

From the data presented in the paper, there were indeed a few fields - four of them to be exact - where non-triazine herbicides were used. Did the researchers compare those to the GM maize fields in order to arrive at their conclusion? No. Possibly because there was no significant difference between the two groups, and in any case, the number of plots was too few to support the claim that the GM maize would be better from the standpoint of biodiversity.

So what did they do instead? The authors noted that on 16 occasions, triazine herbicides had been applied before the maize emerged. On 24 occasions, it had been applied post-emergence only, as had the non-triazine herbicides. They decided to leave out the data from the plots on which triazine was applied pre-emergence, which clearly showed a greater deleterious effect on biodiversity than any other treatment.

The GM maize was thus compared with data from the 24 plots on which the now banned triazine herbicides had been used post-emergence plus the 4 on which non-triazine herbicides were used. There was now a significant difference, which allowed the lead researcher, Perry, to say on the BBC Radio 4 Today Programme that about one-third of the benefits of GM maize would remain after triazines are no longer used.

But this is a highly misleading claim, because 24 out of the 28 plots compared with the GM maize had in fact been sprayed with the banned triazines.

They write in the paper, "If this pooled category of herbicide regimes is indeed representative of weed control in post-triazine conventional crops, and if the weed management in GMHT maize remains the same as observed with the FSE, then final weed numbers would still be larger in GMHT than in conventional maize." There is nothing that would justify the first half of this statement (in fact the authors themselves point out that the non-triazine herbicides had a consistently smaller effect on biodiversity than triazines) and so the claim in the title of the paper is simply bogus.

In reply to criticisms from the House of Commons Environment Audit Committee reported in the Times newspaper, Les Firbank, one of the authors of the Nature paper and also the coordinator of the FSEs, wrote,

"I find it astonishing that the chairman of the committee should announce that the work is "neither robust nor particularly credible science" within a few hours of its publication in Nature, the most highly acclaimed scientific journal in the world."

We find it astonishing that the paper got past the referees of any respectable journal, let alone "Nature, the most highly acclaimed scientific journal in the world".

But that's not the whole the story. The reason yield is not measured is because, if it were, it would very likely reveal a highly significant difference between the GM maize and non-GM maize fields.

Jean Saunders, a citizen opposing the planting of GM crops, has taken the trouble of photographing her local GM maize trial (see the powerpoint presentation here), documenting the severe stunting of the GM maize crop, delayed flowering, and much smaller and fewer cobs compared with the conventional non-GM maize. This finding is surely a lot more relevant to the farmer than data that the scientists have collected and the spin that they have put on the data to allow commercial approval to go ahead.

Article first published 12/03/04


  1. Perry JN, Firbank LG, Champion GT, et al. Ban on triazine herbicides likely to reduce but not negate relative benefits of GMHT maize cropping. Nature 2004 |doi:10.1038/nature02374|
  2. Letter to the Editor from Les Firbank, Times on line,,8122-1034024,00.html

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a cat?