Click the link in this sentence for an overview from September 28 of the StarLink Corn/Taco Bell Taco Shell story

Has the EPA's assessment of the safety of the StarLink Cry9c protein changed?

No. EPA still considers the protein "not dangerous, but unapproved," according to Jim Ailard of the EPA, speaking to a caucus of senators on Capitol Hill on October 26. The concern still centers around the open question of the protein's potential allergenicity.

Steve Johnson of the EPA noted at the caucus that two independent committees of scientists concluded that while there is no evidence for allergenicity, there is not enough evidence to rule that possibility out. Johnson said the agency believes that if there is a risk, it will be remote and low. But the EPA believes the agency lacks the data needed to fully rule out that possibility.

Will the EPA's assessment of the safety of the protein change?

Possibly. Aventis, the company that makes the StarLink corn, delivered to EPA on October 25 a report providing a further assessment of the potential allergenicity of the Cry9C protein in the corn. Steve Johnson of the EPA said agency staff and other experts from outside the government will rigorously review the report. Johnson said the review will take weeks, but not months.

EPA first approved StarLink for use in feed but not food in May 1998. The allergenicity issue has been pending ever since.

What is the status of the recall of StarLink corn grain?

StarLink corn was grown in 1999 and 2000, so the USDA is accounting for corn from two growing seasons. Keith Pitts of the USDA reported to the Senate caucus that the USDA has impounded 88% of the estimated 80 million bushels of Starlink corn grown by 2600 farmers on 525,000 acres in 2000. This area includes non-StarLink corn grown within 660 feet of the StarLink field, but which is also considered unapproved for humans. The buffer zones areintended as a refuge for pests to slow the possible development of populations of pests resistant to the Cry9c protein. Buffer zones are also intended to keep other types of corn pollinated by StarLink pollen from entering the food supply.

Recall by the Numbers, provided by Keith Pitts, USDA

Those 80 million bushels represent about 0.4% of the total US corn crop for 2000.

Of the remaining 12% of StarLink corn crop, about 9.6 million bushels, farmers have shipped 4.8 million bushels to feedlots, as intended and approved.

USDA believes the remaining 4.8 million bushels have been co-mingled with conventional corn. USDA has tracked the 4.8 million bushels comingled, and has found 3.6 million bushels comingled with 75.6 million bushels of conventional corn.

That leaves 1.2 million bushels of StarLink corn yet to find at 116 corn-handling facilities.

Of the 88% that is impounded on farms, farmers can either sell that corn to the USDA at a 25 cents per bushel premium paid by Aventis, or farmers can feed the corn to livestock.

How long will it take to clear the corn supply of StarLink?

Depending how much co-mingled corn is in storage from the 1999 growing season, and how much co-mingled corn goes unaccounted from the 2000 growing season, it could be several years, according to one estimate at the caucus on October 26.

Are there other recalls?

At least seven food companies are recalling products as a result of finding either the Cry9c gene or the protein in foods. The incident has also raised concerns in Japan regarding the status of corn imported from the US. The US currently does not export corn to the European Union, but the incident has raised doubts internationally about the US system for regulating crops,especially biotech crops.

Has the Cry9c protein been detected in taco shells?

No. What has been detected is the gene (the DNA sequence) that encodes for the protein. The protein, which is the original source of concern because it has not yet been ruled out as a potential food allergen, has not been detected in tacos. However, this distinction was not commonly made in the first weeks after the story broke on September 18. It is possible that the protein is present but in quantities too low to be detected, or that the protein has been changed or destroyed by the process of making the tacos.

Is the presence of the gene for the protein enough to consider the shells legally adulterated?

Yes. Both the Cry9c protein and the gene (the DNA sequence) for the Cry9c protein are regulated as plant-pesticides. If either the protein or the gene are detected at any level in a batch of food or a batch of corn flour or a load of corn kernels, that batch or load is considered unfit for human consumption because it contains an unapproved compound.

Is the protein detectable in kernels of corn or in flour of corn?

Yes. Depending on the level of mixing of StarLink corn with other varieties of corn, tests can detect the protein in kernels or in flour. However, the tests for detecting protein are not as sensitive or as specific as the DNA-fingerprinting by PCR (polymerase chain reaction).

What's next?

At the urging of the EPA, Aventis has asked EPA to withdraw the registration of the StarLink plant-pesticide. That means that even if the EPA finds in the next few weeks that the Cry9c protein is not a potential allergen, it's unlikely that StarLink corn will be grown next summer.

If gene-spliced varieties become commercially unfeasible to farmers, then farmers will shift back to growing varieties genetically modified through other methods. Farmers have rapidly adopted gene-spliced crops, and that generally means farmers benefitted from using them, through better pest or weed control or reduced costs of production or reduced use of pesticide sprays. Whatever those benefits are, they will be foregone.

In its agreement with EPA on the split approval of StarLink, Aventis agreed to assume responsibility for failure to keep StarLink out of the food supply. This case will likely set new legal precedents for liability in the food and agriculture sector.

This case will speed changes underway in the US system of food production, processing and delivery. It will likely speed the development of "identity preservation," systems for keeping different sources of foodstuffs separate.

The regulations and precedents set now will likely affect the availability of new technologies, and the opportunity costs of not using new technologies, over the next decade. This period includes the coming of crops developed using genomics. The question of how safe is safe enough, and the imperative to decide in the absence of certainty, and the basic fairness of comparable scrutiny for comparable risk, remain to be balanced and resolved.

Expect a review by Congress of the US system of regulating biotechnology in specific and the food system in general. Regulations specifically intended to reassure the public, rather than to reduce a risk, will likely be proposed. This may rekindle the debate on whether the use of gene-splicing in a crop variety will suffice to trigger regulations that will not be applied to other varieties modified using other genetic technologies.