48 



ionization. 



PLATZMAN: From data at high energy? 



POLLARD: Well, yes. There is a lot of work. I cannot quote them all 

 from memory, but the first one who comes to mind is Williams and Terroux, 

 who did a lot of work with beta rays for this express purpose. They set the 

 expansion to catch not the separate droplets, which is what C.T.R. Wilson did 

 and has been done since, too, but actually to catch just a blob of primary ioni- 

 zation, and then they counted the number of blobs, and they tell the energy that 

 they themselves worked out, in fact, the density of the gas, etc. , and they know 

 the stopping power. In that way you can estimate the number of electron volts 

 that must go into primary ionization. Other work has been done by Beekman, 

 Corson and Brode. All of this work, which was done between 1930 and 1940, 

 agrees quite well. So it looks as though, if one can argue from gases to the liq- 

 uid-solid combination of radiobiology, one can at least say that the average en- 

 ergy that is developed per release is 110 electron volts. This tells you how 

 many primary ionizations there are. It does not tell you how many primary ex- 

 citations there are, and it does not tell you a thing about the energy, this being 

 an average. It is an average over a very wide range. It is useful in telling you 

 the number of processes that must occur. I have taken the Bethe formula that 

 is in the Bethe article, and you can write this down and calculate on this basis, 

 but unfortunately you don't know where to chop it off at the lower end. Start off 

 at 5 electron volts and chop off at 10 or at 15 and you will get a very different 

 result according to where you chop it off. So that is why I was waiting for some 

 information. 



TOBIAS: I found some reports which approached the problem of delta rays 

 in a somewhat different way from what is discussed here. Among others Symon 

 (18) and Igo et al (19), have performed experiments in which they measured the 

 fluctuation in ionization along tracks of particles, in a proportional counter. 

 The fluctuations chiefly arise from secondaries, that is delta rays produced 

 along the track with a distribution of energies. The ionization pulses collected 

 and measured give a statistical pulse size distribution which is different from 

 Poisson's distribution. The shape of the distribution can be predicted on the 

 basis of theories of Landau and others concerning the energy distribution of 

 secondary electrons. Now accurate measurements of this kind could perhaps 

 be used in the future to test the theories of delta ray energy distribution. It may 

 be interesting to some of you to consider for example the pulse height distribu- 

 tion obtained by Igo, et al, on protons. (See Fig. 1.) 



PLATZMAN: How many primary events are there? 



TOBIAS: Each proton makes about 100 secondary electrons in its passage 

 through the proportional counter. 



FANO: Experiments of this type do not contribute important information 

 regarding the energy distribution of secondary electrons. It is known that this 

 distribution follows a l/E 2 law for E much larger than the binding energy of the 

 atomic electrons. The fluctuations of ionization observed in the experiment and 

 considered by the Landau theory just discussed arise from collisions with values 

 of E sufficiently large for the l/E 2 law to hold. Hence any verification of the 

 Landau theory merely confirms what we know but contributes nothing to what we 

 don't know, namely to the distribution of small energy losses. 



TOBIAS: I assumed that at present you have to make assumptions even in 

 the first part of the delta ray distribution, where the delta ray energy is low, 



