The curves which represent the results at 100 meters illustrate the softening of the spectral composition of the radiation with height. The *°k spectrum at 1 meter is quite a bit harder than the 7°"u and ?°*tTh spectra but as the detector height is increased this difference becomes smaller since the higher energy sources in the 23° and *°* th series dominate resulting in a lesser overall rate of softening as compared to that for the single 1.46 MeV *°x source. Figure 4 indicates, however, that the fraction of the exposure rate due to photons of less than 200 keV can become quite Significant as the detector height is increased, depending on the source. Since many detectors used in the field have an energy dependence different from that of air at low energies, it may be inappropriate to use the same calibration factors for these instruments at different heights above the interface. This would be especially true for an instrument calibrated in terms of photon number per unit time such as a scintillation counter since a large increase in the exposure rate due to low energy photons would result in an even larger increase in the actual number of photons at that height. (This can be seen by dividing the points in Figure 2 by energy to obtain the differential number flux curves). The softening of the natural emitter y-ray energy spectrum with detector height has been experimentally verified qualitatively in the field both by Gustafson et al”? and by ourselves in similar field experiments carried out in 1965 using NaI(Tl) detectors. D. , Differential Angular Exposure Rates The differential angular exposure rates at the detector for three different source energies at h < 1 meter, 100 pert meters, and 300 meters are shown in Figure 5. These curves are given in terms of exposure rate per radian due to photons traveling in the direction 8, normalized to a total exposure rate of 1.0, where @ is the angle relative to the