I was very interested in Tom Culliney's information on toxicity of fluvalinate (Hi, Tom!). As most of you would be aware, fluvalinate is not only used in Apistan; it is also an insecticide which is used in horticulture and agriculture on a range of pests, including aphids, thrips, whitefly, and leafroller. It has a residual effect, while at the same time being "soft" on many predator and pollinating insects. It does, however, have a high toxicity on predator mites. In New Zealand, fluvalinate is approved for use on kiwifruit for leafroller, up to and including flowering, but not after. Because it is so "bee safe", its use has greatly reduced bee kills during the early stages of pollination of the crop (when adjacent orchards come into bloom at different times). The Kiwifruit Pollination Association, a beekeepers' group in New Zealand, recommends the chemical to kiwifruit growers. The problem, though, is at the end of the bloom period. Growers here must use other, less "bee safe" insecticides at this time because (I have been told) no tolerance levels have been established for fluvalinate on food. Since almost all New Zealand kiwifruit is exported, and chemical residues of any kind are a major issue in maintaining access to overseas markets, growers here are strictly audited to ensure that fluvalinate is not applied after petal fall. As a result, because of the staggered bloom period between kiwifruit orchards, we still experience bee deaths at the end of flowering. As you can imagine, beekeepers in New Zealand would dearly love to have fluvalinate registered for post-blossom on kiwifruit. As Tom points out, fluvainate has low mammalian acute toxicity (LD50 - 5150mg/kg oral, >2100mg/kg dermal) and no apparent oncogenic, teratogenic, or adverse reproductive effects. So the mystery is why haven't tolerance levels been established for the chemical? Does anyone have any further information on why tolerance levels in foods have not been established for fluvalinate? Is it just a matter of the costs involved in establishing such tolerances, or is there some other issue involved?