H. A. Bumstead — Atmospheric Radio-activity. 7 



Although the differences are not large, they are apparently 

 systematic and indicate that, between 1 and 2 hours, the air 

 activity decays a little more rapidly than a combination of 

 excited activities due to radium and thorium, while between 

 2 and 4 hours it decays more slowly. This cannot be due to 

 the error committed in assuming the thorium activity to be 

 decaying at its final rate ; Rutherford has shown that, after a 

 short exposure, thorium-excited activity at first increases so 

 that a correction for this would be in the wrong direction. 

 Another possible explanation of the discrepancy is that it may 

 be due to an error in the determination of the curve for 

 radium ; a repetition of the determination, however, does not 

 show any increase in the rate between 1 and 2 hours. If, 

 therefore, we do not regard these differences as due to acci- 

 dental errors (which seems unlikely in view of their apparent 

 systematic character), they would seem to indicate the presence 

 of a small proportion of some form of excited radio-activity 

 decaying at a more rapid rate than that which the radium 

 activity shows between 1 and 2 hours.* The present experi- 

 ments are not sufficiently accurate to do more than indicate 

 this as, to some extent, probable ; but I am not without hope 

 that it may be possible to go further in this direction by 

 means of careful experiments especially directed to this end. 

 The only other known induced activity, besides those due to 

 radium and thorium, is that due to actinium. According to a 

 recent determination by Debierne,f its decay is exponential 

 and reaches half-value in 40 minutes. Its rate is, therefore, 

 too slow to serve as an explanation of the discrepancy ; but if a 

 substance of more rapidly decaying activity is present, a small 

 amount of actinium-excited activity is not impossible since 

 there might be partial compensation between the two. 



It is rather remarkable that, although the activity of the air 

 wire is certainly due to a number of different forms of activity, 

 decaying at different rates, the resultant effect between 0*5 and 

 2 hours is so nearly exponential, as indicated by the straight- 

 ness of the lines in fig. 2. It is a further coincidence that, in 

 this portion of the curve, the rate is so near the value found 

 by Debierne for the excited activity due to actinium. The 

 half -value time during this interval is, for Curve I, 38 minutes, 

 for Curve III, 41 minutes. If the observations were not 

 extended further, it would be a natural conclusion that the 

 phenomenon was due to actinium emanation in the air. But 

 the known presence of the thorium activity negatives the sup- 

 position that any considerable part of the effect can be due to 



* Not necessarily more rapid than the final (half-value in 28 minutes) rate 

 of the radium activity, 

 f Debierne, C. E., Feb. 5, 1904, p. 411. 



