A REVISION OF THE ATOMIC WEIGHT OF STRONTIUM. 



SECOND PAPER. — THE ANALYSIS OF STROXTIC CHLORIDE. 



By Theodore William Piichards. 



Presented January 11, 1905. Received January 7, 1905. 



About ten years ago the first paper from this laboratory upon the sub- 

 ject of the atomic weight of strontium* showed that the usually accepted 

 value 87.5 was probably about 0.2 per cent too low. This result was 

 not a surprise, for the error had been predicted, upon the basis of a 

 discussion of the faulty methods used by previous experimenters. The 

 new results seemed to be trustworthy, and to represent, as well as a 

 single series of experiments could, the true value of the constant sought. 



The determination of an atomic weight by a single method is never- 

 theless unsatisfactory, no matter how admirable that method may be. 

 Of course the verdict of a good method and a series of good analyses 

 is much reinforced by an experimental proof of the flaws in other dis- 

 crepant data ; but the greatest of certainty is to be obtained only when 

 the subject has been approached from several wholly different points 

 of view. With this idea in mind, therefore, the analysis of strontic 

 chloride was undertaken in order to confirm or disprove, as the -case 

 might be, tlie results of the analysis of strontic bromide described in 

 1894. The investigation was made in 1895-6, but has remained unpub- 

 lished because of an incomprehensible discrepancy, now explained. 



The balances and weights employed were those used in previous work, 

 to which the reader is referred for all details. All weights were reduced 

 to the vacuum standard as usual. The specific gravity of strontic chlo- 

 ride taken for this purpose was that given by Schroeder, 3.05, which was 

 assumed to be correct, because his determination for baric chloride under 

 similar conditions has often been confirmed. All the care taken in 

 previous work at this laboratory was exercised in this case ; it is unneces- 

 sary to repeat the several details. 



The sample of strontic chloride used for analysis was made from a 

 portion of the strontic carbonate remaining from the third preparation 



* These Proceedings, 30, 369 (1894). 



