292 Messrs. F, Soddy and T. D. Mackenzie on the 



or to change into radium. Meyer and von Schweidler ( Wien. 

 Anzeiger, 1906, xii. Sitzung April 26) have shown that such 

 an extremely slight residual activity does exist, less than 

 1/10,000 part of the initial excited activity, and the present 

 experiment was tried to see if it is, or if it produces, radium, 

 looked for as always by its characteristic emanation. One 

 peculiarity this residual activity exhibits is that it decays 

 rapidly, with a period of 11*7 days, whichisdifficulttheoretically 

 to account for. This experiment was carried out in a different 

 laboratory to the others. 0\1 gram of actinium of activity 

 about 300 times uranium was all that was available, and the 

 first attempt to collect the residual activity from this met with 

 an accident. A second was started on January 2nd, 1907. 

 A loop of platinum wire mounted in glass tubes in a rubber 

 cork, and held by the latter in a small brass cylinder con- 

 taining the actinium in a platinum dish. The cylinder was 

 connected to the positive pole and the wire to the negative 

 pole of the 250 volt supply. The cylinder was closed air- 

 tight, and the experiment left for 160 days, when the wire 

 was removed and examined. It was immediately inserted in 

 a glass cylinder provided with two tubes prepared for it, 

 the whole being closed air-tight. 



After twenty-four hours the air was blown out and tested 

 for the presence of the radium emanation in the electroscope, 

 but no detectable amount was present. The wire was removed, 

 and the residual activity in an electroscope produced a leak 

 of 4*6. The wire was immediately returned to its glass case 

 and left a week till June 9th ; 1907, when it was again tested 

 for the generation of radium emanation. 



This time the wire was heated to dull redness in the dark 

 without removal from its glass case by means of a current. 

 Again not the slightest appreciable leak of the electroscope 

 occurred, although under the conditions of the test the presence 

 of 2 x 10 -12 gram of radium could have been detected. The 

 residual activity of the wire at this stage was 2*9, and eight 

 days later 1'56, which agrees roughly with Meyer and von 

 Schweidler's statement. The wire has been again carefully 

 closed up and will be tested at intervals for the generation 

 of radium. 



In connexion with this problem the results obtained by 

 Boltwood ('Nature/ Nov. 15th, 1906, p. 54 ; Am. Journ. 

 Sci. Dec. 1906, p. 537) and Rutherford ('Nature,' Jan. 17th, 

 1907, p. 270 ; June 6th, 1907, p. 126) may be outlined. 

 Boltwood separated the actinium present in a kilogram of 

 carnotite as completely as possible and sealed it in a glass 

 flask. In the course of 194 days the quantity of radium 



