The most controversial aspect of Bt corn isn't about the hybrids themselves. It's about the non-Bt refuge areas aimed at preventing resistant borers.

University and seed company entomologists all say it's crucial that Bt growers plant part of their acreage to non-Bt hybrids to protect the technology. And most growers say they're willing to do it. But that's where the agreement ends.

Seed companies recommend different-sized refuges, confusing some growers as to what actually is needed. And EPA appears ready to mandate bigger refuges than some companies and growers believe are necessary.

The increased refuge requirements may take effect next year, a year earlier than government-mandated refuges had been expected.

When EPA approved the first Bt corn hybrids in 1996, it asked the seed companies to implement resistance-management plans - including refuge areas - by 2001. Most companies included refuge requirements in their Bt grower agreements from the outset. And most were similar to the refuge plans implemented by Monsanto, which licenses its YieldGard Bt technology to several companies.

Under Monsanto's grower agreement, a grower may plant just 5% of his corn acres to a non-Bt hybrid. But that acreage can't be sprayed with an insecticide if there's a borer outbreak. Or, he can plant 20% of his corn acres to a conventional hybrid and spray it with a non-Bt insecticide if borers threaten yield.

Those acreage requirements are lower than what's recommended by NC205, a group of university entomologists in the North Central Region. They want 20-30% of each grower's corn acreage to be non-Bt hybrids if those acres won't be sprayed; 40% if they'll be sprayed with a non-Bt insecticide.

NC205 member Rick Hellmich, a USDA-ARS entomologist at Ames, IA, says several basic genetic and biological questions must be answered before the optimal refuge size can be determined. These include determining how rare resistant genes are, and how far corn borer moths fly before mating.

He expects that most farmers will understand the concerns and will choose to be good stewards of the technology.

"If we're going to make a mistake, we want it to be on the side of being conservative," says Hellmich. "Economic modeling results suggest that, in the long run, a farmer is better off to plant too much refuge than not enough. If you lose the technology, it's going to hurt you a lot more than the small gains you can get by misusing it."

Novartis Seeds, which sells YieldGard hybrids and others with another Bt gene, recommends that its customers voluntarily plant a 20% unsprayed refuge. EPA seems ready to follow the NC205 guidelines as well.

The first evidence of EPA's intentions came when the agency approved registration of AgrEvo's Starlink Bt corn earlier this year. That registration calls for a 25% unsprayed refuge or a 40% sprayed refuge.

EPA also asked all companies selling Bt corn to submit refuge plans by August. And, at press time, the industry was expecting an EPA notice proposing bigger refuges for all Bt hybrids, starting next year.

That would give needed consistency to refuge sizes. But it also would reduce the economic advantage of Bt corn for many growers.

In September, the National Corn Growers Association (NCGA) expressed that concern.

"We believe in the need for a refuge and will strongly encourage our members to plant one," said Ryland Utlaut, outgoing NCGA president. "But we fear if the refuge requirements are too onerous, growers will not be able to justify using the technology from an economic perspective."

Few observers believe EPA will propose bigger refuges than those outlined by NC205. But Monsanto officials believe those percentages are too high. For next year, the company has increased its unsprayed refuge recommendation from 5% to 10%, leaving the sprayed-refuge minimum at 20%.

"We think the 5% we had initially is still protective," says Monsanto spokesman Dan Holman. "But we know that adoption has increased - there are more growers using Bt than when the plan was put in place. Because of that, we think it's practical to increase the size of the unsprayed refuge."

Holman cites 1997 company research at 153 sites as evidence that a small refuge is sufficient. Virtually no live corn borers were found on plants containing the Bt gene, he says.

Furthermore, he says a survey by Harvest Research, St. Louis, revealed that growers will abandon Bt corn if required refuge areas are too big.

According to Holman, "86% said they would accept a 10% unsprayed refuge, but the numbers drop rapidly after that. And when you talk about a sprayed refuge up to 40%, there are very few growers who see any economic advantage to that. They'll spray their farms instead of using the technology."

That may be a temporary problem, however. Several companies are working to develop hybrids with two Bt genes.

"Our strategy is to come up with products that have lower refuge requirements, which is possible by using hybrids that have more than one Bt gene in them that bind to different sites," says AgrEvo's Keith Newhouse. "They're being tested in the field this year, and I think we'll see commercial products that contain two Bt genes as soon as the year 2000."

Adds Holman: "We think the refuge requirements should definitely be lowered when these stacked products come out. And it concerns us that they're going to a higher refuge now and then will lower it when the stacked products come. We want to make sure we get the right-sized refuge to begin with."