attempting to obtain realistic estimates of population exposure to environmental radiation. The Health and Safety Laboratory team conducted im situ measurements of the radiation field, both out of doors and within residences, to provide a general radiation profile over the populated area, and weighted these results by estimates of mean occupancy time in the various types of locations. This method was first utilized on a large scale by Spiers et al. in Great Britain®-’, and another investigation of this type was carried out in Switzerland in 1961 (ref. 8). The Harvard group undertook a more direct approach and distributed integrating dosimeters to representative members of the population to be worn on the person. For the purpose of comparing the two sets of results, we have defined ‘population exposure’ as the free air dose ‘experienced’ by typical individuals in their usual rounds of activity. We consider only the air dose from the more penetrating components of the environmental radiation field, that is, the terrestrial y- and ionizing cosmic radiation. This definition can be justified by the fact that the mean y-doses to the skin and at various depths in the body are fairly well-known fractions of the free air y-dose®19, and only the more penetrating components of the external radiation field significantly contribute to the dose at the depths of greatest interest, that is, those of the gonads and bone. It should be noted that the possibly important tissue dose contribution from cosmic ray neutrons is not determined by the techniques used in these investigations. The Health and Safety Laboratory radiation survey was carried out in July and August 1962, and subsequent check measurements were made in May and September 1963 and May 1964. The instrumentation included highpressure argon ionization chambers for total dose-rate measurements, a y-spectrometer system for determination of component dose rates (particularly necessary for dis- crimination between the natural and fall-out y-radiation), and portable scintillation detectors for surveys of the areas surrounding each outdoor measurement location as well as for the indoor measurements. The survey techniques, described in detail elsewhere!-1%, provide an overall accuracy of +5 per cent (S.D.) for the measured total dose-rate values and approximately +10 per cent for each of the various components of the total radiation field. The outdoor readings were taken in large, flat open spaces situated in populated areas (for example, parks, fields, lawns, vacant lots) with the instruments placed 3 ft. above the ground. The number of measurements in each area was determined by its population and size, the availability of proper sites, and the observed range and pattern of the previous readings. A sufficient number was taken to ensure that a reasonable radiation profile could be constructed for each area. The quite limited 2