A Quantitative Description of Latent Injury from Ionizing Radiation 335 



any case putting k from equation (8) equal to unity, as is done in equation (6), 

 may modify the constants A and a. This possibility should be considered when 

 comparing the numerical values of these constants in different species and 

 considering their absolute values. 



The constants ft and clJA of equation (1) and their variations with age, if 

 any, can be determined directly. According to equation (1) the injury /, when 

 exposure is stopped, should be repaired exponentially between its initial value 

 and its irreversible residual. This repair was first studied in mammals by 

 Hagen and Simmons (8) using the rat. They assumed exponential repair to 



Radiotion 

 Injury 



AGE S So 



Fig. 3. A schematic representation of the LD50, or lethal injury threshold, for an 

 individual animal with a representation of injury from a single exposure. Data 

 discussed in the text indicate that the irreversible injury is the same, and remains 

 constant independent of the adult age at which it is laid down. This threshold 

 curve cannot be measured directly owing to the change by death of the sample as 

 the age of selection is made older and older. The related curve which is 

 measurable is that of LD50 as a function of remaining life expectancy. 



zero. Usually in making this determination a single substantial sub-lethal 

 dose is given to a large group of animals followed by test doses to sub-groups 

 at increasing intervals. Actually, the residual injury should be determined 

 separately by testing one group after all repair presumably has taken place as 

 illustrated in Fig. 3. The test doses less the residual in roentgens should then 

 demonstrate simple exponential repair according to equation (1). However 

 there is still another complicating factor in that it has been demonstrated recently 

 that all parts of the animal do not recover at the same rate. Carsten and 

 NooNAN (9), for example, have shown that following irradiation of the rat 

 abdomen alone, recovery occurs with a half time between one and two days 

 whereas it occurs in the animal with abdomen shielded with a half-time of 

 about five days, and in the whole animal in about one week according to 

 Hagen and Simmons (8). Possibly, however, the strain used by Carsten and 

 Noonan would demonstrate whole-body recovery at the same rate as abdomen- 

 shielded recovery. In any case we are confronted with the fact that partial 

 body recovery, at least for some tissues, is different from that for whole body. 

 We do not yet know whether recovery of the parts is independent of whether 

 or not other parts have been irradiated. It is fairly certain, however, that 

 faster abdominal recovery can occur following whole-body irradiation, and 

 probably accounts for the several observations (10, 11, 12) that recovery 

 during the first day, as measured by whole-body test doses, is considerably 

 faster than backward exponential extrapolation of later recovery. Strictly 

 then equation (1), in some cases, if not in all, should be written with two or 

 more constants [> and the data analysed appropriately. 



It will be seen that the existence of more than one constant, /i, will not alter 

 the form of equations (4) to (6) but the apparent value of ft determined from 



