210 BELL SYSTEM TECHNICAL JOURNAL 



It is therefore necessary to distinguish, in mind if not in fact, 

 between the "primary ionization" consisting of the electrons and the 

 molecules torn apart from each other by the direct immediate action 

 of the traversing particle,^ and the "entire ionization" (sometimes 

 called "probable ionization") consisting of these together with all the 

 ions formed by the directly-ejected electrons. Under ideal conditions 

 it is presumed that the measure of the former would be the total 

 number of droplet-clusters,^ the measure of the latter would be the 

 total number of droplets, in unit length of path. Not many physicists 

 have tried to evaluate both of these numbers. Of those who have, 

 the data have been scanty, but the consensus of opinion is that the 

 latter is about or not quite twice as great as the former. It is, however 

 likely that the value of the ratio of the two is not important when 

 one wants only to distinguish between electron and mesotron, as we 

 shall presently see. 



The problem of the primary ionization is one of the major tasks of 

 theoretical physics. Classical and quantal theorists alike have spent 

 great labor on the question : given a charged particle of specified charge 

 and mass and speed traversing air (or any other gas), how many 

 electrons will it set free from the molecules in unit length of path? 

 At this point I will give only one of the results — or rather, something 

 which is not a result at all, but a part of the assumptions. It is 

 assumed that as the traversing charged particle flies along through or 

 close to a molecule, it operates upon the electrons thereof by virtue of 

 the ordinary electric forces between its charge and the charges of the 

 electrons. It follows, then, that whatever expression finally may he 

 derived for the primary ionization must depend only upon the charge and 

 the speed of the traversing particle, and not upon its mass. (Mass and 

 momentum of the particle must indeed be great enough to hold it on 

 a sensibly straight course as it plows onward through the gas, despite 

 its losses of energy as it detaches electrons; but this condition is 

 always realized, with the corpuscles of the cosmic rays.) 



I seem to have said that the primary ionization gives no power of 

 distinguishing between an electron on the one hand, a particle of equal 

 charge and different mass on the other. However, it does confer on 

 us this power, for the reason that the curvature of a particle-track in 

 a known magnetic field is a measure not of particle-speed but of 



^ Unluckily called "secondary ionization" by some of the German theorists. 



^ Best to observe the droplet clusters as individual entities, one would wish the 

 expansion to occur before the ions have any time at all to diffuse. To attain this, 

 Williams and Pickup caused the chamber to expand at moments taken at random, 

 and trusted to luck for the appearance of cosmic-ray tracks formed at just the right 

 instants. Luck served them with no fewer than four tracks betokening particles of 

 a distinctive mass. 



