(Reprinted from Nature, Vol. 206, No. May 15, 1965) 4985, pp. 658-662, MEASUREMENT OF THE EXPOSURE OF HUMAN POPULATIONS TO ENVIRONMENTAL RADIATION By WAYNE M. LOWDER and WILLIAM J. CONDON ‘a Health and Safety Laboratory, U.S. Atomic Energy Commission, New York HE accurate determination of representative exposure-levels of large human populations to ionizing radiation in the environment has proved to be a problem of considerable interest to the biologist and of comparable difficulty for the physicist. In an attempt to evaluato existing techniques for obtaining such information, the Health and Safety Laboratory and the Harvard School of Public Health in 1962 undertook concurrent investigations of population exposure to environmental radiation: in selected areas of the States of Vermont and New Hampshire using two independent methods. These ‘investigations have been discussed by Segall’ and by Lowder et al.?, and the extensive results are presented in detail in more recent reports?:+. In this article, we directly compare the two sets of population exposure measurements, discuss briefly some of the alternative methods available for such surveys, and present some gencral conclusions relating to the state of the art which can be derived from our experience in the New England work. The results given here, which partially supersede the preliminary results reported previously!?, provide a useful background for considering the general problems associated with making such measurements and interpreting them properly. The areas chosen for investigation (see refs. 1-4) contain a considerable proportion of the population of the two states, including the major urban centres. Interest in these areas was stimulated initially by the fact that the various underlying bedrock formations appear to differ widely in mean content of naturally occurring radionuclides, as estimated by either direct field and laboratory sample radiometry or inferences from information on similar formations elsewhere®. It seemed possible that these differences in mean bedrock radioactivity might be reflected in significant differences in mean radiation exposure between the populations of these arcas. Entirely unrelated approaches were utilized by the Health and Safety Laboratory and Harvard groups in