CARCINOGENESIS BY IONIZING RADIATIONS 1147 



cells and if a single ionization event can cause a carcinogenic change, the 

 carcinogenic action should be independent of dose rate and dependent on 

 a minimum dose, and X radiation will be more effective than a particles 

 or neutrons. On the other hand, if several ionization events are necessary 

 for the carcinogenic change, the action of the radiation might depend on 

 the dose rate as concerns particles of low specific ionization (such as hard 

 X or 7 radiation), but would be independent of dose rate for heavily 

 ionizing particles. This may apply to irradiation of small volumes of 

 highly susceptible cells, e.g., the induction of bone tumors by localized 

 a emitters. The situation becomes complicated, however, after whole- 

 body irradiation when the systemic secondary reactions caused by the 

 sum of the primary effects are dominant over the primary injury, as may 

 be the case with induction of leukemia or ovarian tumors. 



In order to arrive at a quantitative relationship, the roentgen as a dose 

 unit is adequate for irradiation with 7 or X radiation up to a few Mev, 

 and enables a correlation of data for whole- or part-body irradiation, 

 acute fractionated, or long-continued irradiations. When dealing with 

 corpuscular reactions, the roentgen is of little value, especially if the 

 radiation is localized and the particles have path lengths in tissues rang- 

 ing from several n to several mm. However, the number of particles 

 absorbed and their average energy can be determined, and this energy can 

 be converted to rep. The energy absorbed in 1 g of tissue by 1 r being 

 approximately 93 ergs, the unit rep (roentgen equivalent physical) is 

 defined as that amount of absorbed energy of corpuscular radiation that 

 equals 93 ergs per g of tissue (Lea, 1947). Specific ionization does not 

 enter into the determination of the tissue dose in rep. Since data 

 on the relative efficiency of radiations of different specific ionizations on 

 mammals are scant, but striking differences are observed on small 

 biologic objects, the unit of rem (roentgen equivalent mammal) has been 

 introduced. One rem equals 1 rep divided by an arbitrary factor that 

 expresses the greater biologic effectiveness of corpuscular radiation of 

 high ion density over X or 7 radiation, e.g., as concerns fast neutrons this 



L rep 

 factor is estimated to be about 5. Hence, 1 rem = — ^ — = 0.2 rep. 



This brief outline of the physical aspects of ionizing radiations points to 

 difficulties encountered in obtaining quantitative data on the carcinogenic 

 effects. While the problem of obtaining such data for X or 7 radiation 

 is relatively simple, comparison of such data with those from radiations 

 of different specific ionizations is exceedingly difficult, because a uniform 

 tissue dose cannot be obtained easily by total-body irradiation with 

 corpuscular radiation. When bone tumors are produced by X radiation, 

 the determination of the tissue dose is simple. Bone tumors are more 

 easily produced by /3 or a emitters. The amount of radioactive material 

 deposited in the bone can be measured, but no quantitative comparison 

 can be made with X radiation on the basis of an average computed dose, 



