92 BIOLOGICAL EFFECTS OF RADIATION 



path of the molecules is very short, thermal agitation does not disturb 

 appreciably the localized character of the ionization and therefore the 

 rate of recombination of the ions must depend almost entirely on the 

 linear distribution of the ions in the loci and not appreciably on the num- 

 ber of loci present in a given volume. Accordingly, the rate of recom- 

 bination should be sensibly independent of the intensity of radiation 

 within rather wide limits.^ 



It is possible that the rate of recombination of ions influences the 

 chemical, and hence the biological, effectiveness of a given ionization. 

 Conceivably the more rapidly recombination takes place, the larger will 

 be the proportion of the total number of ions produced per unit time, 

 which would return to the initial electrically neutral state without under- 

 going a rearrangement of the atoms in the molecule. The settlement of 

 this question must be left to future experimentation. At first one might 

 think that the established validity of the Bunsen-Roscoe law (biologically 

 stated, equal effects are produced by equal doses in which intensity X time 

 of irradiation = constant, within limits of time and intensity) in certain 

 chemical reactions, brought about by radiation, precluded any appreciable 

 influence of the rate of ion recombination on the chemical action of 

 radiation. However, as already pointed out, the rate of recombination of 

 ions produced in a liquid or solid must be governed largely by the linear 

 distribution of the ions in the ionization loci, and not appreciably by the 

 number of loci per unit volume (which is the factor that depends on the 

 intensity of radiation). It might be well to remember also that 

 the validity of the Bunsen-Roscoe law in the X-ray region has been 

 established only for a few chemical reactions within rather narrow limits 

 of radiation intensity (3). Exceptions have also been noted (5). From 

 the foregoing viewpoint, one would not expect a breakdown of the 

 Bunsen-Roscoe law until the intensity of radiation became so high that 

 the ion density throughout the material approached that obtaining 

 in the ionization loci themselves. This point has not been tested 

 experimentally. 



The localized character of the ionization produced in a biological 

 material by X-rays has certain important theoretical implications. For 

 this reason it may be well to pursue it a little further. In particular, it is 

 desirable to study the influence of the wave-length of radiation on the 

 ionization loci. 



2 In support of this idea may be mentioned the well-known difficulty of obtaining 

 saturation current in a gas ionized by alpha rays, even when the average ionization 

 is very low, on account of the great concentration of ions along the alpha-ray paths. 

 It is also known that saturation current cannot be obtained in the case of liquids, 

 irrespective of the type of radiation which produces the ionization. 



