A Quantitative Description of Latent Injury from Ionizing Radiation 



333 



it is convenient to measure Sq and S from the beginning of irradiation so that 

 t in equation (5) is replaceable by 5* to give 





a 



/^ 



+ OI.S 



(6) 



This equation represents existing (1, 2) data on chronic irradiation of 

 mammals well within their possible errors. Such errors may be large in long- 

 term experiments owing to infections and other accidents. An example of the 

 fit is given in Fig. 1 for the data in Table I. 



IT 

 bJ 

 a. 



K 



cc 



tli 

 a. 



o 



I- 

 tr 

 o 



X 



in 



10 20 30 40 



SURVIVAL TIME WEEKS 



50 



Fig. 1. Data by Henshaw (17) on chronic irradiation of mice plotted according 

 to equation (6). The data are given in Table I. They cover a wider range than 

 most. The scatter in such experiments frequently increases as the duration of the 



experiment gets long. 



Table I. 



Data by Henshaw (17) on Mice Irradiated Chronical/}' Five Days 

 per Week until Death 



It is obvious now, however, that this equation should fail, providing all the 

 other postulates are valid, because of the inaccuracy of postulate (e) even in the 

 region of adult ages. Actually this postulate could have been written: 'LD50 

 diminishes in proportion to life expectancy.' Consequently it can be tested 

 directly by measuring LDjo as a function of life expectancy. 



In Fig. 2 are plotted LD50 data on Rochester rats (6) as a function of age. 

 It will be seen that LD50 increases with age in young animals, is maximal in 

 young adults, then declines slowly with age. As was mentioned above, postulate 



