OP ARTS AND SCIENCES. 3 



In the next year H. Struve * found that 100 parts of baric chloride 

 produced 112.094 parts of baric sulphate, a value which leads to an 

 atomic weight of barium equal to 137.0. T. Andrews f obtained 

 in 1852 the value 137.6, but he gives none of his details. Six 

 years afterwards Marignac J redetermined the ratio of baric chlo- 

 ride to the sulphate, with a result very different from those of his 

 predecessors. In his hands one hundred parts of the former salt 

 yielded only 112.011 parts of the latter, instead of 112.09 or more. 

 In the same investigation he determined the amount of water of 

 crystallization in baric chloride, with results so unsatisfactory that 

 the values calculated from the various ratios varied from 128.5 to 

 over 138, § as well as the ratio of baric chloride to metallic silver. 

 This last determination led to a value for barium only four one- 

 hundredths of a unit higher than his previous work, ten years be- 

 fore. He admits that the substances used in the analysis were not 

 perfectly pure, but assumes that the impurities were not great 

 enough seriously to influence the result. At about the same time 

 Dumas || was also determining the ratio of baric chloride to silver. 

 He fused the salt in a stream of hydrochloric acid gas, but gives 

 no proof that a slight excess of the gas was not absorbed. If this 

 had been the case, of course the observed atomic weight of barium 

 would have been too low. As a matter of fact, he obtained 137.0 

 for the value of this "apparently variable constant." Below is 

 tabulated a list of the various determinations, grouped according to 

 the processes employed. 



The Atomic Weight of Barium.^ 



0== 16.000. 



Analysis of Baric Carbonate : 



Berzelius, 1811 Ba = 134 to 143 

 Wollaston and Klaproth, 1814 139.2 



Salvetat, 1843 136 



* Liebig's Annalen, 1851, LXXX. 204. 

 t Brit. Assoc. Report, 1852, Part 2, p. 33. 

 X Liebig's Annalen, CVI. 165. 



§ See Meyer and Seubert's " Atomgewichte," p. 176. 

 II Liebig's Annalen, CXIII. 22. 



1 The writer is much indebted to the works of Becker, Clarke, Meyer and 

 Seubert, and Ostwald for valuable assistance in preparing this list. 



