Panel (SAP 1998) as an amount that is 25 times the amount needed to kill 99% of susceptible insects. Empirical information on development of insect genetic adaptation to Bt toxins indicates that such a high dose will kill most partially adapted insects in a pest population (such as heterozygotes). The result is interruption of the typical stepwise process of evolving from susceptible populations, to one that is partially adapted, to fully adapted to the GMPP plant. An analogy can be made to the use of antibiotics to treat human pathogens. The utility of decreasing the survival of partially adapted human pathogens has long been recognized by medical researchers and physicians who routinely recommend that their patients continue to take antibiotics past the period when most of the infectious organisms have been killed. The prolonged treatment period ensures that partially adapted target pathogens will also be killed and so not be transmitted to other people.
Many researchers have examined field and laboratory insect populations in an attempt to understand the mechanisms of insect adaptation to Bt toxins (Tabashnik 1994; Gould and Tabashnik 1998). Results indicate that either multiple genes or single recessive genes are needed to confer full adaptation to a high dose of Bt although partial resistance can be confered by a dominant gene. A very low proportion (1 out of a million, to 1 out of 1000) of fully adapted insects is expected to exist in a population before the population is exposed to Bt. If all host plants on a farm produced the high dose of Bt, only those few insects with the right gene combinations would survive. If they then mated with each other, their offspring would be fully adapted, and the pest population would no longer be affected by the GMPP crop. The planting of nontoxic host plants (refuges) is designed to make sure that a relatively large number of totally susceptible insects are produced on each farm, compared with the few fully adapted insects produced. As long as this refuge is maintained, almost all fully adapted insects produced in the Bt crop are expected to mate with susceptible insects. The offspring of these matings will not have the proper combination of genes needed to be fully adapted, so the evolutionary process is again interrupted. Many researchers expect use of the refuge in combination with plants that produce a high dose of the toxin to increase the time needed for insect adaptation by a factor of 10 (for example, from eight years to more than 80 years) if properly implemented (Gould 1998; Roush 1997; Tabashnik 1994).
By using transgenic or conventional pest-protected plants that contain high levels of multiple toxins with high doses (approach 2), the chance that insects will have the proper gene combination to be fully adapted is further decreased compared to the case where only one toxin is produced by the plant. It also increases the efficiency with which refuge-produced insects can break up combinations of resistance genes from the few pests