CARCINOGENESIS BY IONIZING RADIATIONS 1163 



gists than among other physicians; between 1935 and 1944, according to 

 him, 3.9 per cent of 205 radiologists died of leukemia as compared to 0.44 

 per cent of 34,626 physicians other than radiologists. According to 

 March (1944, 1950) the greatest excess in the mortality of physicians 

 was found from leukemia; the death rate from this disease was 175 per 

 cent of that for white males in the general population. The increased 

 incidence in leukemia among physicians other than radiologists was 133 

 per cent of that of the general population. The incidence of leukemia in 

 radiologists was over ten times as great when compared with other 

 physicians, and this difference according to March was statistically 

 significant. 



Analyzing the conditions of irradiation under which leukemia developed 

 in radiologists and was induced in mice, the following differences are 

 noted: Chronic irradiation produces leukemia in mice only when young 

 animals are exposed over the entire body at a dose rate far exceeding 

 that now encountered by radiologists. Radioscopy involves exposure of 

 only part of the body and such exposures in mice are not leukemogenic. 

 Furthermore, radiologists are exposed to a relatively small average tissue 

 dose. This may explain why the increase of leukemia incidence among 

 radiologists is relatively slight. 



Explosions of atomic bombs may cause leukemia predominantly by 

 virtue of 7 radiation of high energy, and the results in man (Folley et al., 

 1952) and experimental animals (Upton et al., unpublished data) are 

 similar. In both, there is ample evidence that a massive instantaneous 

 single exposure increases the leukemia incidence. In Hiroshima and 

 Nagasaki the increase in incidence of leukemia during the past three years 

 is highly significant in subjects exposed at distances less than 2000 meters 

 and the magnitude of increase is inversely related to the distance from 

 the hypocenter, as indicated by the following very approximate figures: 



Exposure Distance from Hypocenter, m Death Rate from Leukemia per 10 6 Living 

 Up to 1000 500 



1000-1500 210 



1500-2000 87 



Over 2000 32 



Leukemia Deaths per 10 4 Total Dead 

 In all Japan (1943-1949) 13 



United States (1940) 39 



The disease occurred mostly in the early and intermediate age groups and 

 the predominant types were acute myelocytic leukemias (Folley et al., 

 1952). 



Experimental Induction of Leukemia. Carefully controlled experi- 

 mental studies with mice exposed to radiations of an atomic bomb con- 

 firm the preceding observations. Ten months after exposure over 6 per 

 cent of a large surviving population of mice exposed to a dose above the 



