E-mail this article to
yourself or a friend.
Enter address:


Some results of UK biotech crop trials surprised scientists

(Tuesday, Oct. 21, 2003 -- CropChoice news) -- Paul Brown, The Guardian, 10/16/03: The farm scale trials were the largest and most thorough of their kind in the world. Scientists had never previously been able to observe how changing farm practices are affecting wildlife across the country. They cost 5m and lasted four years.

The trials were designed to test whether weeds and insects fared better in fields of conventional crops or crops which had been genetically altered to be resistant to a single herbicide.

In GM crops it meant that the farmer could use one application of herbicide to kill a large spread of weeds in one go without harming the crops. Conventional crops might need several applications of different herbicides at different stages in order to keep weeds under control.

The trials were held because there had already been a steady decline since the 60s in the number of weeds because of intensive agriculture. As a result, there had been a reduction in a wide range of animal species, including bumblebees, grey partridges and corn buntings. They were losing both their food sources and their habitats.

Scientists were surprised to find considerable differences between conventional and GM crops and that they were so marked - as much as five to one in the number of weed seeds produced in conventional oil-seed rape compared with the GM variety. The results were also remarkably consistent across England, Scotland and Wales, although scientists had expected regional variations. This led them to believe the results would apply across the whole of Europe.

There were 273 field trials, 68 fields of maize, 67 of spring oil-seed rape and 66 of beet, both for sugar and fodder. Studies on winter oil-seed rape are still to come.

The GM maize and rape were resistant to Liberty (glufosinate-ammonium) made by BayerCropScience, and the GM beet to Roundup (glyphosate) made by Monsanto. Each trial field was divided into two, half sown with the GM crop and half with its conventional equivalent. Farmers were allowed to treat the crops as they would normally, deciding when to plough and when to plant crops, and when to treat with herbicides.

The researchers monitored the plants and animals in the fields, around the ploughed edges of the fields, before, during and after the crops were grown. Each field was visited 15 to 20 times a year.

Researchers measured the number of grasses and broad-leaved weeds and calculated the weight of the dried weeds. This gave a good measure of the quantity of foliage, flowers and stems that were above ground and available for animals to eat, as well as how many seeds the weeds produced. Another measure was how many seeds fell from the weeds on to the soil surface, known as "seed rain". This allows scientists to predict how many seeds would be available for insects and birds to eat. This is particularly important because some farmland birds - skylark, corn bunting and yellow hammer - which rely on weed seeds in the autumn and winter have been declining. The number of weed seeds left to provide plants for the future was also measured.

The researchers monitored the numbers of insects in and around crops including butterflies, bees, ground beetles, springtails (which live in the soil), and true bugs (which eat other bugs), as well as spiders.



In spring the density of weed seedlings in the GM beet fields was four times that in the conventional crops because many farmers had sprayed to kill weeds in conventional crops before the beet had emerged. However, applying Roundup to the GM crops in May halved the weed density compared with conventional crops. After this the biomass of the remaining weeds was six times lower and the "seed rain" was three times lower compared with conventional crops.

Although there were never many bees and butterflies in beet crops, there were even fewer in the GM beet crops, probably because there were fewer flowering weeds to attract them. There were also fewer butterflies in the tilled margins. Bee numbers, generally low everywhere, were even lower in the GM crops. Growing GM beet is likely to affect populations of weeds in the long term as seed stores will shrink, and will be unlikely to recover.

Spring oil-seed rape

There was 70% less volume of weeds in GM crops and 80% fewer broad-leaved weed seeds. Springtails were significantly more abundant in GM crops in July, and spiders in August, just before the harvest. This was probably because the springtails feed on rotting weeds, which were more abundant in GM crops late in the year. The GM herbicides are used later in the year so the weeds are bigger when they are killed, providing more food for springtails. The spiders were probably feeding on the springtails.


Both the density and size of broadleaved weeds was three times higher in the GM maize fields than in conventional maize fields. Taken together the weeds in the GM crops produced twice as many seeds as the weeds in the conventional crops. Over the growing season butterflies were attracted to the GM maize fields and field margins in the same numbers as conventional fields. There were three times as many honeybees in the GM field boundaries because of more flowering plants, but researches stress that even in GM fields numbers were low. Insects were found in similar numbers in both.

Growing GM maize would be an option for farmers wanting to replace more intensive and persistent herbicides such as atrazine, which is being phased out as too toxic. More weeds and seeds were produced in GM fields, suggesting that birds as well as small mammals like mice might benefit.



The trials found fewer weeds in spring rape and beet fields and butterfly numbers were significantly lower. In the short term butterflies could move elsewhere to find plants to feed on but in the longer term the effect might be important. There were one third fewer butterflies in GM beet crops in July than in conventional crops and in GM spring rape it was half.

Farmland birds

Numbers of birds were not studied directly but farmland birds rely heavily on weed seeds for survival, especially over the winter. For GM beet, weed seeds were reduced by 70% and for GM oil-seed 80%. In addition, for GM spring rape the reduction in seed meant that while the seedbank (the number of seeds in the soil) doubled following conventional crops, it did not increase at all following GM crops. This suggests that GM spring rape plantings will make worse the long term decline in plants, some of which will be important to the diets of farmland birds.

The results for GM maize were the opposite with twice as much "seed rain" on the GM halves of the field.


These thrive where there are weeds in grassland and uncultivated farmland but are generally low in arable fields. There were no differences in the numbers of bumblebees on margins in conventional or GM crops for beet maize or spring rape. While numbers are unlikely to be affected by a lack of weeds in any one year, because they will search elsewhere for food, in the longer term it could have an impact by seriously depleting weed populations within fields. Any reduction in long-term weed numbers could exacerbate the current decline in bumblebee populations in the UK.


One species of seed-eating ground beetle, Harpalus rufipes, is common in arable fields all over the UK. The beetles flourished in conventional spring rape and beet crops later in the season when weeds were producing lots of seeds, but did less well in the GM equivalents because there were fewer weeds and seeds. However, the researchers observed more beetles in GM, rather than conventional, maize crops. If GM crops were grown more extensively than conventional crops the effect on this beetle could be significant over years.