Threshold versus Nonthreshold Response. If chronic effects of radiation are threshold phenomena, 100 pc of Sr® per gram of Ca must be looked upon as a true maximum permissible level and not as an average for large segments of the population. If chronic effects are nonthreshold phenomena and linear with dose (Fig. 8), the maximum permissible levelof sr” in the bones may be expressed in terms of a population or group average, and a portion of the natural population incidence of the effect in question must be attributed to natural background radiation. In this case, the potential hazard should be established on the basis of probability of risk averaged over the entire population or group. 100 r rr ee eee e eee e eee e e 90 F— —_ 80 }— = 70 -— _ EFFECT, % THRESHOLD 60 -—- —4 50 — 40 }— — 30 -- — NON-THRESHOLD 20 -— —_ 10 — olen] I | Ct DOSE, RELATIVE UNITS Fig. 8—Threshold and nonthreshold response as a function of radiation dose. At present it is impossible to @@y whether leukemogenic and sarcogenic responses to chronic radiation dosage are threshold or nonthreshold relationships. The recent Congressional Hearings! failed to produce any deg¥#e of unanimity of opinion among the experts. Argument for a linear relationship between incidence of leukemia and radiation dose was presented recently by Lewis.” His argument was based on all major sources of human data and included a consideration of the Japanese atomic bomb survivors, the British cases of X-ray treated spondylitis, X-ray treated cases of thymic enlargement, practicing radiologists, and spontaneous incidence of leukemia in Brooklyn, New York. The validity of his conclusion was questioned by Warren, Brues, and others during the Congressional Hearings. Radiation as a carcinogenic agent has been discussed at length by Brues,®* who stated that therelation between radiation dose and carcinogenic effect is not easy to find and a critical experiment has yet to be done which will clearly indicate, even-im a single instance, what the relation is over more than a small range of dosages. While admitting that it is not known, he proposes that a threshold relationship between radiation dose and tumor incidence does exist.°7 Without adequate scientific basis but for the purpose of presenting the worst possible potential hazard from Sr® biospheric contamination, a comparison may be made between the radiation dose from present Sr®’ bone levels and the postulated leukemia doubling dose.” Assuming a nonthreshold response and that 10 per cent of the natural incidence of leukemia in the population (6/100,000) is a result of natural background radiation, the average maximum Sr® equilibrium bone level for the north temperate population belt would be equivalent to about 302