SUCCKSSION OF CHANCKS IN KAIMn.M TIVK lioDIKS. '201 



stuiid lor several days, in order to allow all the remaining emanation to escape. The 

 inside of the glass tul>e was found to show considerable activity. The tube was then 

 filled with dilute sulphuric acid, and. after standing several hours, the acid was 

 removed and evaporated down to dry ness in a flat glass dish. Most of the active 

 matter in the glass tube was dissolved out by the acid, and, after evaporation, a 

 strongly active residue was obtained in the glass dish. The active matter was found 

 to give out both a and ft rays, and the ft rays were present in a very unusual 

 proportion ; thus, compared with the intensity of the rays, the ft rays were present 

 in at least 10 times greater proportion than for a thin layer of uranium, thorium, or 

 radium, in a state of radioactive equilibrium. 



The intensity of the a radiation was first tested when the active deposit was alxnit 

 2 months old. The activity, at that time, did not vary apparently over a week's 

 interval. Owing to a numerical error it was at first thought that the activity did 

 not change during a further interval of 3 months ; but the corrected result showed 

 that the activity had more than doubled itself during that interval. Still later 

 olwervations show that the a-ray activity is steadily increasing. 



The increase of the a-ray activity with time has been confirmed by observations of 

 the residual activity left behind on a platinum plate exposed to the emanation. The 

 results then showed that the a-ray activity during the first month, after removal, 

 increases considerably. The relative proportion of ft to at rays steadily diminishes 

 with the time. This is not due to a diminution of the /J-ray activity, but to an 

 increase of the a-ray activity. Further olwervations are in progress to examine the 

 variations of the activity over long periods of time. 



The results, so far as they have gone, show that the residual activity produced on 

 ))odies by exposure to the radium emanation is very complicated. The results 

 discussed in the next section show that the large ft-ray activity is due to matter of 

 different chemical properties from that which gives rise to the a rays. 



The increase of the a-ray activity with time indicates that the deposited matter 

 undergoes a slow ' rayless ' change. The evidence at present obtained points to the 

 conclusion that the deposited matter is initially complex. A small portion of the 

 total amount undergoes a change, accompanied by the emission of ft rays alone. The 

 main portion of the deposited matter undergoes a very slow ' rayless ' transformation, 

 and the resulting product or products give rise to a rays. 



Olwervations extending over a long period of time will be required to determine 

 the period of these changes. It seems probable that radium C breaks up into two 

 distinct products. The major part of the product then undergoes a further change or 

 succession of changes. ' 



* [October 10, 1 904. Further experimental work on this subject has led to a modification of the above 



conclusions. It has been found that the results arc best explained on the supposition that radium C gives 



rise to radium I), which in breaking up emits only fi (and probably y) rays. Radium I) produces radium E, 



which breaks up with the emission of only a rays. It has been deduced that D is probably half 



VOL. CCIV. A. 2 D 



