INFLUENCE OF LINEAE ENERGY TRANSFER 327 



of the LD50 (or any other dose producing some standard degree of effect), 

 two or more radiations can be combined in any desired proportions and 

 their quantitative effect can be predicted on the assumption of complete 

 additivity. If then the predicted degree of effect is experimentally 

 observed, it is concluded that the radiations of different LET are indeed 

 completely additive; if the observed effect is ciuantitatively less than 

 that predicted, incomplete additivity is demonstrated. 



Additivity of radiations differing in LET by more than a factor of 10 

 has been investigated for three biological effects: inhibition of hatching 

 of Drosophila eggs (item 37, 13, y, fission fn) ; lethal action on broad bean 

 roots (item 80, X, a) ; lethal action on mice (item 51 ; item 49, y, fission 

 fn). For Drosophila eggs additivity was complete, and for the broad 

 bean it was nearly so. For mice it was complete when the doses were 

 given in about an hour (item 49), but incomplete when irradiation was 

 spread over a day or two (item 51). 



When additivity is incomplete, the radiations involved obviously differ 

 in some qualitative feature of their mechanisms. 



On the other hand, complete additivity does not prove that all radia- 

 tions operate through identical mechanisms. Thus, one radiation may 

 operate through some series of relevant events such as 



Irradiation -^A-^B^-C^'E 



in which the final effect E is determined when a certain number of indi- 

 vidual decisive events of type C have occurred per cell (or other bio- 

 logical unit). Another radiation might operate through the same type 

 of decisive event C, except that this event may be produced by the 

 sequence 



Irradiation ->A-^B'->C^E 



where B' is different from B. In this case the two radiations would add 

 completely, since they operate through the same decisive event C, but 

 their mechanisms are not identical. 



Additivity is discussed in more detail in a recent review (Zirkle, 1950). 



Theoretical Implications of the Observed Influence of Linear Energy 

 Transfer. The mere fact that LET has a quantitative influence on a 

 given radiobiological action proves that at least some of the events com- 

 prising that action occur while the ionization tracks still retain some of 

 their configuration, i.e., before the ions and excited molecules (or other 

 chemical entities derived from them) have time, by diffusion or other 

 means, to become distributed singly at random. This time may be of 

 the order of only 10~^ or 10~^ second (Lea, 1947b). It is rather impres- 

 sive that such a wide variety of radiobiological actions — indeed, all 

 that have been investigated with LET varying more than a factor of 

 10 — should have their quantitative relations even partially determined 



