It's worth wrapping up in a wet towel to marvel at a transparently pure example of scientific scholarship on the sins of epidemiologists with a mission. You can skip the flights of academic mumbo-jumbo - for example, about "grilled hamburgers with a mean BaP content of 9 µg/kg" - and positively revel in the total demolition of the leading 'Research' on ETS and lung cancer.

Among the fatal flaws are: biased selection, which ignores unwelcome findings; changed "confidence intervals", which amounts to fiddling the margins of error; "meta-analysis", which makes an omelette of bad and rotten eggs; "self-reported information" which confuses ex-smokers with non-smokers; "dose response" which ignores amounts of exposure to ETS; "confounding factors", which ignores differences in diet, hereditary disposition, life-style, etc; and "study utility", which judges the value of research by its results not by its quality.

Scientists, seeking funding and recognition by ruling orthodoxy, strive ever-harder to fake figures.

Their incentive is the knowledge that insignificant or negative findings are less likely to attract attention even in scientific journals, let alone make it newsworthy for the popular press.

In short, "passive smoking" is a hoax inspired by anti-smoking pressure groups (ASH, etc), obligingly invented by militant medicos, and unwittingly spread by passive thinking! It is driven by familiar political imperatives and orchestrated by media scare stories. For all the effort to prove their point, the anti-smokers' ETS claims vanish in a puff of smoke.

This book is expensive - 200 cigarettes - but the Nilsson chapter alone is worth the price for any serious smoker, or student of medical myth-making.

Lord Harris of High Cross

• What RISK? edited by Roger Bate, published by Butterworth, 1997, price £30.