MEASUREMENT OF IONIZING RADIATIONS 165 



ments in a layer of air of equivalent thickness and by the use of Eq. (13), 

 where p,„ would refer to the ratio of the stopping power of the object to 

 that of air. To be at all vahd, however, the ionization measurement must 

 be made with chambers properly designed, namely, with walls of mini- 

 mum thickness and light supporting structures. These should be placed 

 as much as possible outside the beam in order to minimize extraneous 

 scattering within the collecting volume. Examples of desirable tech- 

 niques will be given later. 



Status of W and p™. The experimental evidence on the value of W in 

 any one gas as a function of particle mass, charge, and energy has been 

 reviewed in considerable detail by Gray (1944). He concluded that for 

 electrons of energy greater than 0.3 Mev the most probable value of W in 

 air is a constant and is equal to 32.0 ev, and that below 0.3 Mev there 

 seems to be a steady rise in W with diminishing energy until at 1 kev the 

 value is probably 37 volts. This conclusion should be revised in the 

 light of recent experiments with proportional counters reported by Cur- 

 ran, Cockcroft, and Insch (1950) to the effect that W = 31.0 for air for 

 electrons of 5.6 kev mean energy (/3 rays from H^). These authors, more- 

 over, consider it a strong probability that W is a constant for any one gas 

 from 0.5 to 30 kev with the possible exception of O2, which would account 

 for a slight anomaly in air. 



The determination of W in air and in argon as a function of energy, for 

 alpha particles, has been the object of recent studies by Jesse and collab- 

 orators (1950) and a general review by Bethe (1950). Bethe concludes 

 that the bulk of the experimental evidence bearing on the problem dem- 

 onstrates that (1) the ionization in argon is strictly proportional to energy 

 absorbed, namely, that W is constant irrespective of the energy and of the 

 type of the ionizing particle (H^ H^ He^, L-, C^^) ; (2) in view of the 

 earUer experiments of Gurney (1925) W should be considered independent 

 of energy also for the other noble gases; (3) the anomaly of W in air for 

 protons and a particles is to be considered real and that it amounts to 

 about 16 per cent in the a range of to 5 Mev; and (4) there are good 

 reasons to believe that W has the same value in any one gas for either H^ 

 or He"* of the same velocity. In view of the recent experimental methods 

 used, the question might be raised as to whether the anomaly of W in 

 air and O2 is real or caused, either wholly or in part, by the well-known 

 experimental difficulties of obtaining "saturation" conditions with 

 a particles in air and of eliminating the effect of electron attachment 

 to molecular oxygen or methane in proportional counters triggered by 

 electrons. The lack of discussion on this point in the most recent reports 

 leaves the question open, despite the fact that the results obtained in air 

 by Gurney (who has made positive statements on the subject of satura- 

 tion) have been essentially confirmed by Jesse and collaborators. 



Since the ratio pm of the stopping powers of the medium and of the gas 



