the onset was greatly accelerated by the radiation exposure. For doses of less than 100 r in humans andin statistically significant numbers of experimental animals, there are no defini- tive data. The available experimental data from fairly extensive studies at higher levels of radiation suggest that, depending on the type of leukemia, the induction curve may be either sigmoidal or linear. It is not known whether or not there is a threshold for leukemia induction by radiation. While it has been generally accepted among students of leukemia that there is some accumulated dose of radiation, perhaps in the vicinity of 50 r below which leukemia is not induced, Dr. E. B. Lewis of California Institute of Technology and Dr. Hardin Hones of the Uni- versity of California at Berkeley have proposed the hypothesis that leukemia induction from ionizing radiation is a linear function of dose regardless of dose rate and have suggested that for each mr average exposure per year to the entire population of the United States, there would eventually be an additional 10 cases of leukemia per year, i.e., about 40 cases per year from fallout. Using this same reasoning there would be roughly 1200 cases per year as a result of diagnostic medical x-rays. It has also been postulated that bone-seeking radioactive nuclides such as radiostrontium might be leukemogenic. The present average body burden of Sr®° in children in the United States is slightly less than 1/100th the maximum permissible bone concentration for Sr®* for the population as a whole. This has been given as 0.1 yc for an adult, i.e., 100 wuc per gram of Ca. 100 upc per gram of Ca would lead to an exposure to nearby bone marrow of about 0.14 rad per year, that is, about 10 rad in a lifetime or less than 5 rad in 30 years. If Lewis’ hypothesis is correct, that leukemia induction is linear with dose to the bone marrow, and wereall the bone marrow to receive this dose, which it does not, such a body burden for all people in the United States could mean an additional 5 to 10% increase in leukemia (500 to 1000) each year, There are considerable experimental data indicating that with large single doses leukemia does not result if a fair fraction of the hematopoietic system is shielded from total body radiation. With a certain type of mouse lymphoma, even shielding one extremity of the animal will vitiate the leukemogenic effect of a large single exposure to radiation. It appears then that if leukemia in general or even one type of leukemia can be the result simply of a radiation-induced somatic mutation untempered by homeostatic factors, fallout at the present rate of weapons testing could, on the basis of certain assumptions as to the number of cases of leukemia due to background radiation, result in some 30 to 40 additional cases per year or about 1000 per generation. The same assumptionslead to a figure of 1200 cases per year as the result of medical x-rays (12,000 per generation). If small amounts of Sr* relatively uniformly distributed in bone can indeed produce leukemia in the manner postulated by Lewis, fallout from continued weapons testing at the present rate could eventually lead to 35 to 250 additional cases per year (1000 to 7000 per generation). To compiete the story, one must keep in mind that cocarcinogenic factors and additive factors may in certain susceptible individuals prepare the way for a small dose of radiation to trigger a case of leukemia. 5 BONE CANCER The present incidence of bone sarcoma in this country is about 2000 cases annually. It is apparent from the observations of radiotherapists that a dose of something more than 1000 r given locally to the bone is required to induce cancer, and cancer induction by doses of less than 2000 r is a very rare occurrence. As to the induction of cancer by chronic irradiation from bone-seeking radionuclides, we have a considerable body of data in human beings. Briefly, it can be stated that no case has come to light of bone cancer in an individual exposed to “pure” radium salt in adult life who had left in him at the time of observation (usually 20 to 30 years after the material was ingested) less than 0.4 ug of radium plus an undetermined amount of mesothorium. The National Committee on Radiation Protection and the International Commission on Radiological Protection have taken a little less than 1/10th of this figure, 0.1 yc, as the permissible radium burden for adult workers. The correspondingfigure for Sr®® is 1.0 uc. One-tenth of that or 100 puc of sr®° per gram of Ca, the presently considered per- missible body burden for the population as a whole, would give about 0.26 rad per year or 20 rad in 70 years, i.e., approximately three times the exposure to bone from naturally occurring radioactivity. 280