In late July 1995, two chemical company representatives and a farm editor were riding to a field day on a Sumner, Miss., farm that was the location of one of the first Bt cottonseed production fields in the United States.

The conversation turned to the outbreak of pesticide-resistant tobacco budworms that had decimated cotton fields in Mississippi, Alabama and other parts of the Southeast earlier that summer. Some growers didn't have enough cotton left after the budworms came through to warrant a picker in their fields.

“Until this happened, we were beginning to wonder what we would do with all of this new seed,” said one of the representatives. “Now we probably won't have enough to meet the demand.”

The rest, as they say, is history. Farmers in Alabama planted nearly 95 percent of their acres in cotton containing the Bacillus thuringiensis gene in 1996, the first year the cotton, which became known as Bollgard, was commercially available. In other states, growers planted 60 to 75 percent of their fields in Bt varieties.

Bollgard cotton became the most widely adopted new technology since the invention of the metal plow because farmers knew they could plan on spending a set amount of money for controlling tobacco budworms and bollworms and avoid open-ended spraying.

But the technology was not without its challenges. Growers learned early they could improve yields by overspraying Bollgard with pyrethroids. And they began to try to deal with the “refuge” requirements for Bt cotton.

From the beginning, EPA said cotton producers must set aside a portion of their cotton acres for non-transgenic varieties to provide a “reservoir” of susceptible moths that had not been exposed to the Bt toxin. The idea: forestall or slow the development of resistance to the Bollgard technology.

Unlike pesticides, which might be sprayed four or five times a season, the Bt toxin was in the field 24/7. Many entomologists felt it was only a matter of time before budworms and bollworms became tolerant, if not, resistant to the Bt gene.

Computer modeling by entomologists showed tobacco budworms could become resistant to the Bt toxin expressed in the transgenic varieties within 30 generations. Before EPA approved Bollgard for use, it was already working with Monsanto, university scientists and grower groups to develop a resistance management plan.

EPA officials called the insect resistance management program for Bt cotton “unprecedented.

“We felt the technology was so important to farmers that we had to make sure we maintained the ‘shelf-life’ of the Bt toxin as long as possible,” the EPA's Sharlene Matten told a group of scientists attending a USDA Agriculture Research Service-sponsored Heliothis/Helicoverpa Workshop in Dallas in 2003.

The insect resistance management plan involved seven components: (1) structured refuge, (2) resistance monitoring, (3) remedial action plan, (4) compliance assurance, (5) grower education, (6) grower agreements and (7) annual reports.

Farmers quickly became familiar with the structured refuge requirements, which initially included a 96 percent Bollgard/4 percent non-Bt option or an 80 percent Bollgard/20 percent non-Bt option. The 20 percent refuge could be sprayed with insecticides while the 4 percent could not.

EPA subsequently added new wrinkles to the refuge requirements to try to make the non-Bt areas more effective in providing susceptible moths for mating with potentially Bt-resistant moths and to help farmers reduce the yield loss and added insecticide expense of acres planted to non-transgenic varieties.

One of those was a 5 percent “embedded” refuge option that allowed growers to plant at least 5 acres of non-Bt cotton for every 95 acres of Bollgard cotton — or Bollgard II later on. The 5 percent had to be planted as a contiguous block within the Bollgard or Bollgard II field or within the field unit.

The change allowed growers to treat the 5 percent, embedded block whenever the entire associated Bollgard or Bollgard II cotton field or field unit was sprayed with an insecticide. The 5 percent block had to be treated with the same insecticide, except for foliar Bt products, at the same rate and in the same 24-hour period. (Growers could also split the 5 percent refuge among different locations in very large fields (longer or wider than 1 mile).)

On non-Bt acres that were not embedded within a field or field units, growers were prohibited from spraying cotton with any lepidoptera-active insecticide for the control of tobacco budworm, bollworm or pink bollworm from first square through harvest. EPA gave producers a list of the active ingredients of those insecticides.

Allowing sprays prior to first square came about in response to concerns that growers in Louisiana and Texas might lose their stand to insects such as cabbage loopers, which would have defeated the purpose of the refuge acres since bare fields could not serve as hosts to susceptible budworm and bollworm moths.

“Most growers in that area used the 5 percent refuge option, and based on the refuge requirements, they were not allowed to treat that 5 percent of their acreage for cabbage loopers,” said Walt Mullins, Bollgard technical development lead for Monsanto.

“Their concern was that if we don't manage those cabbage loopers there's a risk we could lose the stand in the 5 percent refuge. What that would have meant was that they would have had to replant the refuge, which would have been considerably behind the Bollgard cotton.”

Growers who used a 20 percent sprayed refuge option (80 percent Bt/20 percent non-Bt) could treat the refuge with any insecticides except foliar Bt products. Fields of Bollgard or Bollgard II cotton had to be within 1 mile — or preferably, according to EPA, within 0.5 mile of the associated refuge.

For producers with smaller fields or operations spread over large distances, EPA established a community refuge plan that allowed growers to band together to set aside a non-Bt block of cotton for their refuge acres. Growers in the community were required to use the 20 percent sprayed option, a 5 percent unsprayed option or a combination of the two.

As might be imagined, keeping the refuge acres properly placed and sprayed or not sprayed — as the requirement might be — proved to be a formidable task for many producers. At one point, EPA said, its internal studies showed a 75-percent compliance rate for the refuge requirements.

Shortly after Monsanto introduced Bollgard cotton, it began working on the development of a second generation Bt cotton, the technology that became Bollgard II. The latter contained two Bacillus thuringiensis genes, Cry 1Ab and Cry 2Ab, instead of the one gene that had been genetically engineered in Bollgard cotton.

Dow AgroSciences was also developing new technology for its Phytogen cottonseed that contained a trait that expresses the Cry 1F and Cry1Ac proteins from Bacillus thuringiensis for the control caterpillar insects in cotton.

The development of Bollgard II was another step in Monsanto's portion of the insect resistance management plan. Besides broadening the Bollgard control spectrum to include more insect pests, the two-gene approach was also expected to make it more difficult for budworms and bollworms to become resistant to Bollgard, said Mullins.

“The primary motivation for developing Bollgard II was to enhance the resistance management value of the Bt product,” he said. “It was broadly believed that putting two genes in the plant that worked differently on the insect would reduce the probability of resistant insects surviving both genes and significantly decrease the time to resistance.”

Monsanto also began working with university researchers to try to determine if the broader spectrum and increased efficacy of the two-gene technology could help simplify or at least lower the refuge hoops growers had to jump through to use the transgenic cotton.

One question involved whether the primary insect targets in the Southeast — tobacco budworms and bollworms — had enough natural host plants that growers could stop planting non-Bt cotton as a refuge for susceptible moths.

One of the scientists who worked on the issue was J.R. Bradley, a professor emeritus at North Carolina State University. Bradley and other researchers conducted studies in six cotton-producing states to determine the host range for bollworm and tobacco budworm. The tests used C-3 and C-4 radio-labeled isotopes to determine the insects' food sources.

“The studies indicated that bollworms also have a lot of hosts, including peanuts, soybeans, weedy hosts, corn and sorghum, throughout the season and that they're flying significant distances all season so there is significant mixing of insects,” said Monsanto's Mullins.

After seeking more information about the studies and the concept, EPA last spring approved the use of a natural refuge option for Bollgard II cotton and for Dow's Widestrike Insect Protection trait in the Southeast states and selected counties in Texas.

Under the natural refuge option, growers in the approved region will not have to plant a non-Bt cotton refuge for Bollgard or Widestrike cotton. A structured non-Bt cotton refuge will continue to be required for Bollgard cotton and for Bollgard II and Widestrike cotton planted in other counties in Texas and in Arizona, New Mexico and California.

Some cotton entomologists and agronomists were not happy with the change. They said more studies are needed to make sure enough alternate host plants are available in areas such as the lower Mid-South to prevent resistance from occurring sooner rather than later.

But EPA officials disagreed. “Use of Bollgard II cotton with its two distinct Bt proteins, in conjunction with the use of a natural refuge, will enhance cotton insect resistance management,” they said in a Pesticide Program Update.