The Importance of Being Quantitative When Crying “Fallacy”¹

Abstract

Wartenberg and Gallo(') find fault with Ames et ~l . (~) for using the rank order of rodent TD5,s to predict the rank order of carcinogen hazards to humans when exposures are at much lower levels than the doses typically used in the rodent bioassays. This is certainly a topic that merits serious investigation, but Wartenberg and Gallo do not offer any new perspectives or much analysis of existing information. Rather than provide a thorough investigation of the problem, their sole purpose seems to have been to refute a paper that they apparently did not read carefully enough.The authors call Ames et al. (2) to task for proposing "a new model" in which the rodent TD50 is used as the basis for evaluating human carcinogenic hazards. But the TD50 or a similar high-dose measure of potency is already widely used for risk assessment, and Ames et ~l . (~) are quick to point out that this is an inadequate, if sometimes expedient, approach: