XIV. X RAYS AND X IRRADIATION 487 



variability, even though smaller in amount and often negligible in 

 many experiments, would correspond with the interspecies variablitiy 

 already known. 



The physical effects of radiation and the hypotheses that may 

 explain the effects have been rather well worked out. There are 

 some criteria by which the logic may be tested for its applicability 

 to the experiment. For the one absorption type of exponential 

 X-ray effects the data, when plotted on the arith.-log. grid, will have 

 to be a straight line if the hypothesis is suitable. The accurate deter- 

 mination of such a line is strong evidence that the effects of irradiation 

 are along physical rather than biological lines. It hardly seems prob- 

 able that the biological effect of the irradiations would create such a 

 simple mathematical curve. However, a further means of proving 

 the physical interpretation exists in the fact that, if the dose is frac- 

 tionated instead of given continuously, the same biological inactiva- 

 tion occurs. This should follow on the physical hypothesis but not 

 necessarily on the biological hypothesis of innate variability. A 

 further test of the effect of radiation is to treat the material with 

 radiations of different types. Radiation over the X-ray band should 

 behave in a similar manner due to the fact that the larger quanta 

 yield recoil electrons that, in effect, reduce the short wavelength 

 radiation to the long wavelength type before the production of many 

 ions occurs. This reasoning will not hold for densely ionizing par- 

 ticles, as a rays, which produce distinctly different quantitative re- 

 sults from those of X rays of medium wavelength. These effects 

 would not be expected on general biological grounds. 



The fact that inactivation curves do not behave strictly according 

 to the simple exponential relation is not necessarily evidence against 

 the physical theory that a single absorption act is sufficient to produce 

 the observed biological results. Several factors are significant in 

 X-ray experiments. If the bacteria or viruses should be treated 

 and be clmnped so that it is impossible to get survival values on single 

 organisms, there will be some divergence from the linear relation 

 expected. A second factor that may produce results discrepant 

 from those expected is the shielding of biological material from the 

 radiation by other substances that may surrovmd it. The effect of 

 such shielding could be particularly marked with long wavelength 

 X rays. Another important factor in causing this curvilinearity is 

 that the X rays may create toxic products in the solution irradiated, 

 and these products in turn affect the particles in solution, rendering 



