EFFECTS OF FEEDING CONDITIONS ON TOXICITY FOR THE PURPOSE OF EXTRAPOLATION

1. Mathematical models can be most helpful in attempts to understand the way effects of toxic substances show up under various conditions. They are essential for extrapolation and prediction, especially when effects have to be quantified. 2. In some cases, there is a tight relationship between effects and toxicokinetics, i.e. uptake/elimination behaviour of compounds. 3. It can be shown that, at least for some compounds, the toxicokinetics depends on feeding conditions. This alone makes it necessary to account for energetics in the use of laboratory observations for predictions and interpretation concerning field data. 4. It can also be shown that, given certain effects on individuals, the consequences for populations depend sensitively on energetics. This leads to an even tighter link between ecotoxicology and energetics. 5. A basic problem is that realistic models involve a relatively large number of parameters, even under the most simple assumptions about kinetics and effects. This constraints possibilities of extrapolation and prediction. 6. All in all, a close link between experimental and modelling programmes is necessary.