In the air. Pollen from canola (inset) moved up to 3 kilometers between fields.

A Little Pollen Goes a Long Way

A major concern about genetically modified (GM) crops is that they might spread their genes to nearby weeds or organic crops. Now, a comprehensive study of canola fields reports that pollen can travel considerable distances, but that the amount of gene flow is minimal. Although the findings reinforce the difficulty of growing GM-free crops, they also suggest that the levels of gene diffusion are below European standards for contamination of conventional food.

Over the last decade, a handful of small experiments has indicated that a minuscule amount of pollen from engineered crops can spread up to a few hundred meters. But what happens on real farms was unclear. To find out, a team led by reproductive ecologist Mary Rieger, of the Cooperative Research Center for Australian Weed Management and the University of Adelaide in Australia, took advantage of a unique opportunity. In 2000, Australian farmers for the first time planted varieties of canola with resistance to acetolactate synthase-inhibiting herbicides. Although these crops are not GM varieties--they were created by mutating existing canola genes, not by introducing new ones--their unique DNA should move about just like DNA from GM crops.

Working in three states and under various climatic conditions, Rieger's team collected seeds from 63 nearby fields planted with conventional canola. The herbicide-resistance trait spread to 63% of the conventional fields, including some up to 3 kilometers away from a source. But although the gene traveled far and wide, it got into only a small proportion of conventional plants. When averaged per field, the highest percentage of herbicide resistant seeds was 0.07%. The harvests from the vast majority of fields contained less than 0.03%, the team reports in the 28 June issue of Science. This implies that contamination of non-GM canola would be less than 1%, which is the cutoff that Australian regulators have discussed as acceptable and that their European counterparts have provisionally approved. But the study underlines a clear risk: Once transgenes are introduced, they can't be completely controlled.

With these new results in hand, researchers should have a better handle on gene flow when canola is modified in other ways, says herbicide physiologist Linda Hall of the University of Alberta, Edmonton. Although pollen from any crop should travel in similar ways, Hall and others note that extrapolation is tricky because crops reproduce in different ways. "Canola is one of the more problematic in terms of gene flow," says plant geneticist Rikke Jørgensen of the Riso National Laboratory in Denmark. "This is a worst-case scenario."

Related sites
More articles about gene flow
and herbicide resistance
Mary Rieger's home page