In the highly charged debate over biotech crops, critics and supporters can agree on one thing: New strains are sold and planted before much is known about their ecological effects, and too little is being done to find out about them.
Genetically engineered corn, soybeans, and cotton already are staples on millions of acres of farmland in North America. Yet studies of potential risks often are conducted after a new strain of gene-spliced seeds has taken root on farms - too late for the environment or the approval process.
As a result, some researchers say, the world's fields are becoming a laboratory for the largest unplanned ecological experiment in agricultural history.
"We look at genetically modified crops, and we know what the potential hazards may be," says Alan McHughen, a plant geneticist at the University of Saskatchewan in Saskatoon. "But we want to know what the hazards really are before they go out" into the marketplace.
"The question of hazards is real, and it's not going to go away," he adds.
Two recent studies highlight potential ecological risks of genetically engineered crops.
In the current issue of the journal Science, a pair of British scientists suggest that starling populations could drop by as much as 90 percent if farmers adopt a new strain of sugar beets tailored to tolerate herbicides. Using a computer model, the scientists found that the starlings' plight depended on how widely farmers adopted the sugar beets and how much herbicide they applied to their fields - depriving the birds of the weed seeds they eat.
Ironically, the threat to the birds didn't come from the plants themselves, but from farming practices that could result from the plants' introduction.
Corn and butterflies
Also, late last month, researchers from Iowa State University in Ames published the results of experiments studying genetically modified corn and butterflies. It suggests that monarch butterflies are threatened by pollen from corn that's engineered to produce a toxin fatal to a pest called the European corn borer.
The work, which appeared in the journal Oecologia, is the latest in a series of yes-it-does, no-it-doesn't results from various scientists trying to determine whether the pollen seriously threatens the butterflies.
On a scientific level, "there is nothing terrible about any of these studies," even when results appear contradictory, says Jane Rissler, a senior scientist at the Union of Concerned Scientists. "No one study is really definitive; you need an accumulation of evidence over time."
But researchers say they face a number of hurdles in trying to build that body of knowledge.
First, the regulations governing tests can be counterproductive. Scientists must isolate test fields of key crops such as corn and canola because they produce copious amounts of pollen, which can be blown beyond the field and inadvertently mix with other plant species. Yet offsite effects are among the risks scientists want to quantify.
Moreover, test plots are vulnerable to attacks from activists who oppose any use of genetically engineered crops.
On Aug. 25, a group called the Los Nios del Maiz claimed responsibility for an attack that destroyed 1,500 to 2,000 corn plants and damaged a range of other plants in four greenhouses at the University of California at San Diego's biology field station. Only a small fraction of the plants the group destroyed were the result of genetic engineering, according to Maarten Chrispeels, director of the university's Center for Molecular Agriculture.
In other cases, it can be difficult to repeat experiments. "The amount of transgenic material [genetic information that can be transferred] is limited," says John Obrycki, one of the Iowa State University scientists involved in the most recent corn-pollen study.
Needed: more time for research
Initially, it's possible to work with the seeds' developers to find out whether a plant is likely to pass a trait or toxin to another species. But here in the US, regulators require only a one-year trial before a seed strain is approved for sale. Once the seeds go on sale, Dr. Obrycki says, companies often are less willing to give them to researchers wanting to conduct further tests for potential risks. He argues that the approval process should require two years of testing instead of one.
Finally, funding remains tight for risk studies. In the US, for example, the Department of Agriculture spends about $1.7 million a year for risk assessment.
"That's a tiny amount of money compared with what the biotech industry spends to develop and market new products," says Rebecca Goldburg, a senior scientist at Environmental Defense.
The result, critics say, is an approval process that lacks sufficient data to assess the risk of genetically engineered crops.
That theme also emerges from a generally biotech-friendly report from the National Academy of Sciences. In the April study, the academy noted that, as the proportion of genetically engineered crops and the rate of their introduction increases, more research is needed to assess risks to human health and the environment.
Yet others say more money for risk studies is not the answer. They argue that scientists should mine data already accumulating on biotech crops. Such information can be incorporated into models that can help researchers predict possible environmental outcomes from planting genetically engineered crops.
Yet even data-mining proponents look longingly at a comprehensive research project under way in Britain to address risks from biotech crops. The government is spending 8 million ($11.7 million) over three years to look at the effects of modified corn, canola, and sugar beets.
Says Frank Forcella, a plant geneticist with the University of Minnesota at Morris: "That level of funding for field experiments is fantastic even by Yankee standards."
(c) Copyright 2000. The Christian Science Publishing Society