A series of official tests was recently carried out on the ‘Anderson’ air raid shelter, and the results published. The A.R.P. Co-ordinating Committee, which is an independent body of architects, surveyors, medical men and men of science under the chairmanship of Prof. J. B. S. Haldane, has made some serious criticisms of these tests in a statement forwarded to Sir John Anderson. The Committee admits that a certain amount of useful information has been obtained, but it considers that insufficient tests were carried out to justify what it calls ‘far too optimistic’ conclusions. A number of interesting points are raised; for example, the Committee thinks that it is not enough to do each explosion test once only; and if, as is claimed, apparently similar bombs often behave differently in exploding, the objection would seem to be justified. The Committee also asks for further research into the physiological effects of blast, a subject which is very imperfectly understood. It is well known that ‘blast’ is, in effect, a very rapid rise in pressure followed by a slightly slower fall. It is this rapid fluctuation of pressure, presumably, which gives blast its lethal effect. The Committee rightly points out that there is strong evidence that it is not the magnitude of the changes in pressure which matters so much as the rate at which the pressure changes. The reactions of the human body to fairly slow changes of pressure have long been well known to physiologists, but there is little exact knowledge of the results to be expected when the body is subjected to wide fluctuations in a fraction of a second. If, as wo aro told, the general population is to be exposed to just such changes, the Committee would seem to be justified in calling for more research.