On the Limitations of the Theory of the Positron

sensitivity and, in spite of the raggedness of the trace at this extreme sensitivity, a displacement of the mean of about one centimeter could have been detected. This would be one-two thousandth of the effect measured in Fig. 1 and, since the proton-current used was four times the deuton-current, we may conclude that any effect produced by protons on carbon is less than one-eight thousandth of the effect producl'(d by deutons. [Incidentally, this is additional evidence for the extreme purity of our proton-beam after magnetic analysis, as previously reported.] The wide discrepancy between this result and those reported by the workers in Cambridge and Pasadena does not appear to us to be particularly surprising, since magnetic analysis evidently was not used in their experiments. While it is true that in the experiments at Pasadena residual contamination by deuterium would hardly be expected to amount to 10 percent or 20 percent of the total beam, it is quite possible that the total beam with deuterium flowing might contain only five to ten times as many deutons as with the deuterium-flow cut off. In some of our previous work we determined the yields of several efficient disintegration-processes by working for many days with no deuterium-flow, that is, using only the residue of deuterium left in the tube from previous runs. Further evidence that the percentage of deutons in the Pasadena experiments is much lower than estimated is given by the higher gamm-ray yields from carbon and beryllium we have observed.4 We take great pleasure in expressing our gratitude to Professors Urey, Zanetti, and La Mer, of Columbia University, for the supply of deuterium which made these experiments possible. L. R. HAFSTAD M.A. TUVE Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, D. C., June 1, 1934.