This brief study, by a German medical engineer and an American colleague fed up with the media tendency to lump health-hazards together, has the outward trappings of objectivity. The antidote, in their opinion, is to establish a comparative risk-scale: we could distinguish between the unlikelihood of being killed by licensed drugs and such commonplaces as traffic accidents. Their proposed ""safety-degree scale"" is a logarithmic sequence representing variations in the size of a ""unicohort,"" or the total number of people who can be shown to have been at a specific risk divided by the number of actual victims (say, 5 out of 20 or 500 eaters of bad fish in a restaurant, for a unicohort of 4 or 100). The case seems at best premature for all but a handful of large, well-defined risk-comparisons (e.g., cigarette smoking, by far ""the single biggest proven hazard in today's life,"" is likelier to kill you than any food additive). Still, Urquhart and Hellmann's attempt to achieve some perspective is by no means inapropos. With rather more temperance and flexibility than Edith Efron (The Apocalyptics, p. 183), they summarize the state of several troubling issues like dioxin contamination and exposure to asbestos, remind us of the many epidemiological unknowns in trying to understand trends in the incidence of both cancer and heart disease, and suggest that our general struggle to make sense of ""a new world of omnipresent, strange chemicals"" is ""a process analogous to the sanitation and hygiene which our great-grandparents had to learn."" But their portrayal of ""adventurous, risk-taking behavior"" as the cornerstone of human progress becomes more strident in the discussion of drug regulation, where they argue that many more lives are lost than saved through the slowness of the licensing process (and the first few million purchasers of a newly marketed medicine will inevitably constitute an inadvertent guinea-pig group). Misleading simplifications and glossings-over of problems appear throughout the book: the suggestion that formaldehyde poses only an uncertain ""potential toxicity problem in man"" is quite incorrect (the authors apparently mean carcinogenicity problem). As a corrective to errors of perspective, this has some good moments. But its implication that actuarial numbers-games can clear the communal vision ignores the anything-but-homogeneous distribution of very serious hazards and the rudimentary state of fact-finding.