DOSE-EFFECT CURVES 341 



rated by diffusion. The ions have reacted to form other entities. A few 

 OH radicals remain at this instant, but most of the surviving relevant 

 entities are H2O2 molecules. The ratio of relevant to irrelevant entities 

 is much lower than in the younger (lower) track because many of the 

 potentially relevant entities have been removed by irrelevant processes. 

 The possible diffusion path of one of the H2O2 molecules from the older 

 track to the precursor is indicated. 



Let us now try to generalize, that is, select from our outline of a special 

 case (Fig. 1) those states and processes which may reasonably be ex- 

 pected to be parts of any radiobiological action. These are listed in the 

 skeleton outline of Fig. 2, and the relevant ones are as follows: 



a. Normal state of cell. 

 A. Energy transfer. 

 /3. Activated molecules present. 

 M. Decisive process, 

 ju. Decisive state. 

 CO. End effect. 



Of these we can be absolutely certain that a, A, and co are relevant. We 

 can be certain that state jS occurs, but, although its relevance seems 

 highly probable, we cannot be certain that it is relevant to all radio- 

 biological actions. (The relevant mechanism in some actions may not 

 involve chemical changes at all but work through physical processes 

 entirely, as in Read's proposed mechanism of chromosome breaks.) For 

 no radiobiological action do we know the natures of decisive process M 

 and decisive state ju, but I have included them in the general outline 

 (Fig. 2) because I intend to use them as convenient general concepts, 

 even though it is likely that in many mechanisms the decisive state n 

 is identical with the final state co. 



I shall now briefly consider some of the most prominent radiobiological 

 facts, interpreting them in terms of the foregoing outlines of mechanism. 



Quantitative Aspects of Dose-Effect Curves 



Let us first consider the quantitative relations between dose and ef- 

 fect. We here imply that both the dose and the effect can be measured. 

 I shall assume that the dose measurement is the affair of the physicists 

 and is well in hand. Some remarks are necessary, however, about meas- 

 urement of effect. In the first place, since every radiobiological effect 

 is a deviation from a normal condition, various degrees of effect are al- 

 ways expressed in terms of a datum for a normal control population of 



