A Search for Protons in the Primary Cosmic-Ray Beam

It has been suggested that a large proportion of the cosmic rays found at sea level are protons. Now a charged particle is characterized by the fact that its ionization increases enormously towards the end of its range, so that, in the case of protons and alpha-particles large and measurable spurts of ionization should be produced in relatively short distances by those rays which are ending their journeys. If $r$ is the distance from the end of the range to the point where the ionization per centimeter of path is $\sigma$, then, the fraction of the rays which, passing through a length $l$ of a vessel containing gas at pressure $p$, produce therein spurts of ions greater in number than $\mathrm{lp}\sigma$, is ($1-e^{-\mu r}$), or approximately, $\mu r$, where $\mu$ is the absorption coefficient of the radiation concerned. The assignment of a lower limit to the spurts which can be measured determines $\sigma$, and so $r$, through the aid of Bethe's theory; and, it becomes possible to calculate how many such spurts should be observed if the rays are protons. In an experiment based upon the foregoing principles spurts above an assigned size (7.2×${10}^4$ ion pairs in one experiment, and 3.9×${10}^4$ ion pairs in another experiment) produced in a small ionization chamber were measured. To eliminate alpha-particles the chamber was divided into two halves by horizontal partition, and only spurts occurring simultaneously in both halves were counted. Spurts produced by "showers" were recognized and eliminated by suitably arranged Geiger counters. A conservative interpretation of the results gives an upper limit for the number of protons present as five percent of the total number of cosmic rays at sea level, or 12 percent of the intensity of the hard component if Compton's estimate of the absorption coefficient and intensity of the hard component be used. If Millikan's values are adopted our estimate of the upper limit is six percent.