CONTEMPORARY ADVANCES IN PHYSICS 201 



based on the presumption — often admitted in the context to be more 

 than doubtful — that the particles to which they refer are electrons. 

 The danger might be mitigated by describing these as "quasi-energy- 

 values" expressed in "quasi-Mev." — For actual electrons with mo- 

 menta as great as those figuring in the cosmic rays, the energy-value 

 in electron-volts is practically equal to 300 times the 7/p-value ex- 

 pressed in gauss-centimetres. 



Many a cosmic-ray particle suffers no deflection that can be detected 

 in its entire course across a Wilson chamber (diameter, 15 cm. or 

 even more) in a magnetic field as strong as can be applied over so 

 great a volume (field-strength, 20,000 gauss or thereabouts). One 

 might well be tempted to think such a particle chargeless, for if this 

 were the case, the field would have no grasp at all upon it; but if it 

 were chargeless it could not ionize the molecules of the gas and therefore 

 could not form the chain of ions on which the droplets are founded. 

 In some of the finest of the experiments (those in Pasadena and those 

 in Paris) a detectable curvature of the track would be shown if this 

 were made by an electron of energy so enormous as 2 • 10^" electron- 

 volts (20,000 Mev!). The uncurved tracks accordingly speak of 

 electrons of energies greater than 20,000 Mev, if these particles are 

 electrons; and the inference is not much less drastic, if they are more 

 massive than an electron. 



We, however, are more interested, for the present, in the tracks 

 which are sensibly curved ; and most of all, in the tracks which are inter- 

 sected by a metal plate and which show a curvature on one side of the 

 plate and a larger curvature on the other (Figure 9). From the two 

 p-values one can deduce the momentum-loss Ap and the energy-loss 

 A£ sufifered by the particle in passing through the plate. (Yet I 

 emphasize again that A/» is computable only if the charge is correctly 

 guessed, and A£ only if the rest-mass is correctly guessed in addition 

 to the charge.) With this ambition Anderson inserted such plates 

 for the first time into a Wilson chamber, in 1931. The idea had a 

 wonderful and unforeseen result, some years ago recounted in these 

 pages. Notice that above I spoke of the momentum-Zo55 and the 

 energy-loss sufifered by a particle in going through a plate. In so 

 doing I was making the assumption that it is a loss and not a gain 

 which happens. If this highly plausible assumption is correct, then 

 the sense in which the particle is traveling its path is knowable; it is 

 from the side of the plate on which the curvature is less, to the side 

 on which the curvature is greater. If the sense of the motion is 

 knowable, so also the sign of the charge of the particle is knowable, 

 being positive or negative according as the track is bent with its 



