286 Mr. Norman Campbell on Delta Rays. 



the emergence and incidence radiation could be obtained by 

 taking- a value fcr V = 0, which was not certainly less correct 

 than that given by the mean of the actual measurements. 

 On these grounds I am inclined to conclude that there is no 

 difference of quality between the emergence and incidence 

 radiation, and that the small apparent differences are acci- 

 dental and due to experimental uncertainties. 



11. If this conclusion be accepted, the second part of the 

 investigation, the comparison of the quality of the delta rays 

 for different velocities of the alpha rays, is rendered easier. 

 For, if the two branches of the curve are really similar, we 

 may avoid placing great reliance upon measurements at the 

 zero by simply adding the ordinates PN and P'N X (fig. 1) 

 corresponding to the same difference of potential with a 

 different sign. The results of such addition are given in 

 Table II. Here again the columns 1-9 represent the results 

 of different thicknesses of the layer covering B, and hence, 

 for different velocities of the emergent alpha rays. The 

 numbers in any column represent the ratio of the arithmetical 

 sum of the currents obtained for positive and negative values 

 of V given in the first column to the sum of those values for 

 V = 40. (It appeared throughout that no change in the 

 current occurred on increasing the value of V beyond 40 

 volts.) On the simple theory the numbers in any row should 

 represent the fraction of the delta rays which have a speed 

 less than that corresponding to the value of V in the first 

 column. 



It appears at once that the difference between different 

 columns is very small. I am inclined to think that any 

 apparent differences are due to the residual effect of the 

 sources of error mentioned already : for, if there is an un- 

 avoidable difference of potential v between the electrodes, 

 the measurements for an imposed difference of potential V 

 will not represent the current due to V, but the sum of the 

 currents due to Y — v and V + v. If there is any difference, 

 it is that the values of the current for high velocities of the 

 alpha rays tend to be greater for small values of V and less 

 for large values : that is to say, that there are more rays of 

 very low and of very high speed, and fewer rays of inter- 

 mediate speed. The experiments seem to establish conclu- 

 sively that there is no very marked dependence of the 

 velocity of the delta rays on that of the exciting primary 

 rays : it is not certain that there is any dependence. Such 

 a conclusion, if it is accepted, is obviously of the utmost 

 importance for the theory of ionization by alpha rays. But 



