352 SPECULATIONS ON CELLULAR ACTIONS 



Let us briefly examine this last notion with regard to each of the three 

 types of competition: (a) irrelevant processes, (6) reverse processes, and 

 (c) restitution (Fig. 2). (a) If the absolute rates of any relevant process 

 and of the irrelevant process (es) competing for the same relevant entity 

 are both dependent on the first power of the concentration of the relevant 

 entity, then the relative rates of the relevant and irrelevant processes 

 are unaffected by this concentration and therefore by the time pattern. 

 If a more complicated situation prevails — for example, if the relevant 

 process depends on the square, and the irrelevant on the first power, of 

 the concentration of relevant entity, or vice versa — then one process is 

 favored over the other by a change in time pattern, and a change in h 

 occurs. However, there is no way to predict whether the relevant or 

 irrelevant process will be favored and consequently whether h will be in- 

 creased or decreased, {h) The foregoing statements apply also to the 

 influence of time pattern on the relative rates of relevant and reverse 

 processes, (c) The situation with regard to restitution is perhaps more 

 specific. Restitution is any process (or series of processes) by which the 

 cell tends to regain the initial state, aside from reverse processes in the 

 strict sense used here. Restitution processes would seem to be of three 

 possible types: (1) processes involving no radiation-produced entities; 

 (2) processes involving relevant entities and natural ones; (3) processes 

 involving two or more species of relevant entities. Type 3 would seem 

 to be relatively so improbable that we may neglect it. In processes of 

 type 2 it would appear extremely unlikely that more than one species of 

 relevant entity should be involved, and therefore the reaction rate would 

 depend on the first power of its concentration. If the relevant process 

 involving this entity should depend on a higher power of its concentra- 

 tion, then the effectiveness of the radiation should decrease as the time 

 pattern is prolonged. In restitution of type 1, the rate is completely in- 

 dependent of the time pattern, and accordingly, even if the relevant 

 processes proceed at rates dependent only on the first powers of the con- 

 centrations of relevant entities, the effectiveness of the treatment is de- 

 creased by prolongation of the time pattern. 



In earher paragraphs we have seen that we cannot predict, as the time 

 pattern is lengthened, the direction in w^hich the irrelevant processes or 

 reverse processes are likely to change the effectiveness. On the other 

 hand, we have just seen that restitution is likely, in all its most probable 

 mechanisms, to decrease effectiveness (h) if the time pattern is prolonged. 

 Now we have already noted that, among the large number of radio- 

 biological actions known to be dependent on the time pattern, all but a 

 few decrease in eflficiency with extension of the time pattern. This would 

 seem to be an indication that the influence of time pattern is usually 



