215 TABLE 77. M? ano C ror Varrovus Gases Gas 1. Has 2. CF, Pressurerange, ure C 2-11 1.3-11 5.0 10.3 42.2 84.0 9-27 5-20 4-12 3-14 3-10 3-12 2.3-9 1.9-8 1.1-8 1.0-4 1.5-9 1.9-6 1,0-6 2.414 2.4-18 4-12 5.2 6.2 9.9 10.2 11.9 10.6 14.2 18.4 19.6 17.5 22,0 23,0 25.1 8.0 6.8 3.2 | 3. PHs 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18, 19. CH, CH,OH C.H;OH (CH3)20 (CH3).CO eyeloC3H, (CH;}CH nC (CH3)},C CH, cyeloCgHis nCelT is nC His Xe CsHy H.0 2-12 | 4.6 | | 45.9 53.8 66.4 97.7 105.2 118.0 106.2 141.9 184.8 182.9 162.4 213.2 223.4 256.2 72.4 68.8 32.3 o(em?) = 1.874 % 1072 (M2xr, + Cx») B I a= 1 g ge 1 Bt —1 lxw=- Be (velocity of primaryelectron) /(velocity of light}. results quoted above to be the most reliable, but the uncertainty in the absolute values is not indicated. Limits of error will be discussed in the following section of this report. Our previous measurements on Xe and C,H, were less self-consistent than we would like, so newsets of data for them were obtained and analyzed. The data for H,O are reanalyzed. The results, given in the last three lines of Table 77, supplant the results reported earlier. LIMITS OF ERROR Errors maybest be discussed in relation to the Bethe formula, which has been used as the basis for smoothing and averaging our observations, On theoretical grounds, the formula should be expected to describe accurately the energy dependence of cross sections throughout and well beyond the range of our measurements; within their limits, the measurements conform to this expectation. The formula maybe written o = (1.874 & 10°" em*)C[(M?/C)a, + 2]. Its form is illustrated in Figure 166, where o/C is plotted against log £ for the values of M’/C = 0.08, 0.10, and 0.12. Amongourresults, values of C range from 7 to 250, TABLE78. entrance windowof a gas-filled counter. Calculations showed that at 0.1 MeV the average path of primaries E, MeV between entrance and exit windows is 114 % greater than the distance between the windows; the difference becomesnegilgible for energies above 0.5 MeV. Previ- 200 Theresults for CO, reported previously were obtained by Geiger Mueller counting. When we learned that proportional counting could be used with this gas, a new set of measurements with this mode of counting were carried out to provide a eross check. The results orig- The same data, reevaluated and reduced according to our current criteria and method, yield the values W= 5.754010, C= 559 +04. eee fTRU ang SRE ear The new data, with proportional counting, give the result MM = 5754007, 1.0015 + 0.0041 ‘ Relative standard deviation, based on counted numbers. report. C= 55.7 + 0.6. 0.9960 + 0.0057 1.0075 + 0.0059 Average by increasing AZ” by 4% and decreasing C by 34%; the correction has been applied to all values quoted in this M’ = 6.14 + 04, otfae 0.77 1.10 ously reported results can be corrected for this effect inajly reported, corrected for scattering, are Comparison oF [LONIZATION CROSS SECTIONS OF ELECTRONS AND PosrTRONS ON ARGON w E S 150 ts 100 4 10 8a blo é0 T PRESENT EXPERIMENTS 4 60 6 40 4 30 3 Q C= 579403. The uncertainties indicated are the standard deviations as given by the least squares solutions, each involving about forty data points, and indicate the degree of internal consistency of the data. We believe the last 20 | 105 | 10¢ L 107 { 108 10? 2 ev Fic. 166—-Bethe formula ¢/C as a function of primary en- ergy. &, with 1f°/C as a parameter,