DISCUSSION 



It will be seen that the second postulate resembles that of Lea but is not identical 

 with it. 



Some typical results are given in the table. Allowance has been made for : 



(/) The low energy electrons fS-rays) produced by the more energetic electrons. 



(2) The relative contributions of photoelectric absorption and Compton scattering 

 to the total tissue dose, and the distribution of energy among the recoil electrons. 



(j) The spectral distribution of photon energy from tubes operated at particular 

 kilovoltages and filtrations, where such data were available. Unfortunately no such 

 data are available at present for the radiations used by Kirby-Smith and Daniels. 

 The figures given in columns I and II for lightly and heavily filtered 250 kvp 

 X-radiation therefore refer to the equivalent monochromatic X-radiation. 



On the basis of either postulate, a significant variation of biological efficiency 

 with X-ray quality in the range 50-200 kV is theoretically to be expected. In the 

 case of postulate (a) the biological efficiency falls steeply from unity at 0-5keV to a 

 local minimum value of 0-135 at 29keV, rises to a local maximum of 0-183 at 



Table I. — Relative Effectiveness of X- and Gamma Radiation 



Calculated values of y) compared with experimental data of 

 Kirby-Smith and Daniels. (Chromosome Structural Damage) 

 Column I values of y; for Q_= • 5 keV 

 Column II values of/; for 0-5<Q_<3-0 keV 



85keV, and falls to 0-077 for ®"Co gamma rays. In the case of postulate {b) the 

 minimum value is 0-242 at 34keV, the maximum is 0-40 at 76keV, and the ®°Co 

 gamma ray value is 0-115. Both postulates give a computed ratio of X-ray to 

 gamma ray efficiency somewhat greater than is observed. It is interesting to note, 

 however, that the observed difference of 33 per cent in biological efficiency as between 

 250 kvp filtered by 3 mm Al and the same kilovoltage filtered by 4mm Cu is to 

 be anticipated in terms of postulate {b). Column II of the table in fact shows a 

 computed difference in efficiency distinctly greater than that which was observed, 

 but the computed ratio would probably be somewhat reduced if allowance were 

 made for the spectral distribution of the two X-ray beams. 



S. H. Revell : I quite agree with Gray that the exchange hypothesis which I 

 have suggested is in one respect very like the version of the breakage-and-reunion 

 hypothesis evolved by Catcheside and Lea. I also think that there is probably no 

 point in spending too much time considering whether or not this exchange hypothesis 



279 



