Decembee 4, 1908] 



SCIENCE 



807 



emanation from 0.3Y g. of radium. Tlie 

 quantity of emanation which was effectively 

 destroyed in the apparatus was a little less 

 and was equivalent to the saturated emana- 

 tion from 0.27 g. of radium. When the ex- 

 periment was considered finished, the solution 

 was transferred from the apparatus into a 

 platinum crucible and a few drops of nitric 

 acid added. Into this same crucible we 

 dipped a piece of platinum foil on which we 

 deposited the copper. When the solution was 

 freed from copper, it was evaporated to dry- 

 ness in the crucible and the latter was heated 

 just enough to drive off the sulphuric acid. 

 The residue was dissolved in a few drops of 

 water and treated with hydrogen sulphide to 

 remove the last traces of copper still remain- 

 ing. The liquid was filtered through a 

 platinum filter into a platinum watch-glass of 

 known weight and was evaporated to drsmess 

 at a very moderate temperature. The minute 

 residue was weighed. 



The same treatment was applied to 7 c.c. of 

 a copper sulphate solution which had not been 

 subjected to the action of radium. The final 

 residues were examined spectroseopically. 

 Their weights were 0.0004 g. and 0.0005 g. in 

 the real experiments, 0.0003 g. and 0.0002 g. 

 in the blanks. It is to be noted that the 

 amount of copper taken is very close to that 

 used by Mr. Eamsay. The amount of emana- 

 tion consumed is also approximately the same 

 (1.85 mm.' in Mr. Ramsay's phraseology). 

 In spite of this, the residue finally obtained is 

 much less. 



The spectroscopic examination showed that 

 the residue consisted chiefly of sodium with a 

 little potassium ; the presence of lithium could 

 not be established. An experiment with a 

 mixture of sodium and lithium sulphates 

 showed that we could detect, though only with 

 difficulty, the presence of one part of lithium 

 sulphate in ten thousand parts of sodium 

 sulphate. With the same ray it was easy to 

 detect one part of lithium sulphate in three 

 thousand parts of sodium sulphate. Con- 

 sequently the amount of the metal lithium 

 which could be present was less than 0.6 X 

 10-° milligrams. 



With the same amounts of copper and of 



emanation, Messrs. Ramsay and Cameron 

 found 1.7 X 10~* milligrams of lithium. If 

 there is an error in calculation and this num- 

 ber refers to lithimn chloride, there would 

 stiU be 3 X lO"" milligrams of metallic 

 lithium. 



The residue which we obtained was in each 

 case much less than that obtained by Messrs. 

 Ramsay and Cameron and this difference is 

 probably due to our not using glass vessels. 

 The difference in the weights of the residues 

 obtained by us in the real experiments and 

 the blanks is very small (0.1 to 0.3 milligram). 

 The difference is probably due to the introduc- 

 tion of traces of foreign substances into the 

 apparatus along with the emanation. In the 

 most reliable experiment of Messrs. Ramsay 

 and Cameron, the same difference is 0.88 milli- 

 gram and we believe that this is due to the 

 solution attacking the glass more vigorously 

 in presence of the emanation. 



The following check experiment was made. 

 Into a copper sulphate solution containing 

 0.27 g. of copper we introduced an amount of 

 lithium, sulphate corresponding to 1.7 X 10~* 

 milligrams of LiCl; this solution was then 

 treated in the same way as in the other blank 

 experiments. In the residue finally obtained 

 it was very easy to see the red ray of lithium. 

 This shows that the lithium was not elimi- 

 nated by the treatment adopted. 



To sum up, we must say that we have not 

 succeeded in confirming the experiments of 

 Messrs. Ramsay and Cameron. It is evi- 

 dently impossible for us to say that no trace 

 of sodium or lithium is formed during the ex- 

 periment. We believe, however, that the 

 formation of these elements can not be con- 

 sidered as an established fact. 



SPECIAL ARTICLES 



MOMENTUM EFFECTS IN ELECTRIC DISCHARGE 



In the issue of July 17 a partial report was 

 made of experiments on electric discharge 

 around a right angle in a wire. Since then 

 some of the methods have been modified, and 

 additional results have been reached. 



In all of the work thus far, an eight-plate 

 static machine has been used. It may be 



