BEIR IV report.

In the Research News article by Richard A. Kerr "In search of elusive little comets" (10 June, p. 1403), the position held by John Craven of the University of Iowa was incorrectly given. He is a senior research physicist.

Neutrons at Hiroshima: how their disappearance affected risk estimates.

Since the 1960s, follow-up studies of the atomic bomb survivors have been the major source of quantitative information on the late effects of radiation on human health. Prior to 1986 these observations were misinterpreted in that the dose equivalent for survivors within 1600 m at Hiroshima (the proximal Hiroshima survivors) was thought to be due mainly to neutrons rather than to gamma radiation. The 1986 binational reassessment of atomic bomb dosimetry in Hiroshima and Nagasaki showed that neutrons made a minor contribution to the dose equivalent in both cities. The virtual elimination of neutrons as a causative factor for observed excess cancers among A-bomb survivors has had a profound effect on radiation risk estimates. Neither the high risks nor the linear dose response observed among proximal survivors at Hiroshima could be attributed to neutrons. Consequently, a linear-quadratic response pattern for the induction of solid cancers by gamma radiation is no longer supported by the Japanese experience. In this paper we explore why the neutron flux at Hiroshima was overestimated in the past, what was learned in the binational dose reassessment for Hiroshima and Nagasaki, and how this affected the recent risk assessment by the BEIR V committee.