ABSTRACT
The problem of risk assessment would be easier if one could realistically recommend open informed debate of the issues of major risk–benefit analyses. Dioxins are an example of toxins where species differences appear to be of enormous importance in evaluating likely risk, where orders of magnitude differences in sensitivity are encountered and where humans appear to be very much less sensitive than might have been supposed from animal experiment. This chapter provides two points relating to selection of animal species: in initial toxicity experiments one has no basis for choosing mice rather than elephants, and one species of animal does not predict for another species more closely related than is man to the investigated species. The high-dose exposure situation would seem to offer the best chance for the successful exercise of extrapolative toxicology, and its numerous failures must give all cause for concern about the infinitely more difficult task of the environmental toxicologist.
