Assuming primary electrons, the theoretical variation ofshower counting rate with altitude has been computed and a comparisonmade with the observations of Kraybill. Expressed asthe ratio to sea level, the observed counting rate has a maximumof 63 as compared to 24 for the theoretical curve. The observedmaximum is near 27,000 feet while the computed maximumis at 22,000 feet. Approximations in the calculation and observationaluncertainties may make this difference in the elevationsof the maxima spurious. The approximations are such as togive too low a theoretical counting rate, but an estimate of the magnitude seems to leave the theoretical maximum counting ratetoo small by a factor of 1.5 to 2. A computation based upon amechanism for the multiple production of secondaries by primaryprotons, which has been proposed by Lewis, Oppenheimer and Wouthuysen,would probably lead to better agreement with the observations.
The zenith angle distribution of showers detected by Kraybill’scounters at 30,000 feet has been computed and comparedwith one determined by Mr. E.W. Cowan from cloud chamber measurements. The theoretical distribution is twice as broad as the observedone. The angular distribution inferred from the altitudedependence observed by Kraybill is in much better agreement withthe theoretical than with the cloud chamber distribution. Thediscrepancy of the cloud chamber observations may arise from thefact that it was only possible to assign a direction to the tracksin 20 to 30 per cent of the photographs.
