M. Stas's New Determination of the Atomic Weights. 139 



sor for the atomic weights. The original memoir must be con- 

 sulted for an account of the great labour and pains taken to 

 ensure the purity of the substances used, accuracy in the weigh- 

 ings, and the exclusion of all sources of error from the apparatus 

 employed. In these respects the memoir is probably one of the 

 most important which has ever appeared. In one point his de- 

 terminations differ from those of preceding chemists — that is, in 

 the quantities taken, which are very considerably larger than 

 those usually employed ; the balances used were on a correspond- 

 ing scale, and of the most perfect construction. It may be 

 mentioned that he found any kind of glass attacked when heated 

 directly by flame ; but that, when protected by a coating of char- 

 coal or of magnesia, it could be heated to softening without un- 

 dergoing any alteration in weight. In all cases, where practi- 

 cable, Stas used vessels of platinum. 



The present communication refers to the atomic weights of 

 nitrogen, chlorine, sulphur, potassium, sodium, and silver, and 

 their relation to the atomic weight of hydrogen. The author 

 made for this purpose the synthesis of the following substances — 

 chloride of silver, sulphide of silver, nitrate of silver, nitrate of 

 lead, sulphate of lead ; the analysis of chlorate of potash, and 

 sulphate of silver ; and he determined the relations between the 

 atomic weights of the following substances: — silver and chloride 

 of potassium, silver and chloride of ammonium, silver and chlo- 

 ride of sodium, nitrate of silver and chloride of potassium, nitrate 

 of silver and chloride of ammonium. 



A most essential point was the preparation of pure metallic 

 silver. Several different methods were used, one of which con- 

 sisted in digesting a dilute solution of silver with finely divided 

 phosphorus. This action is slow, but the metal, after having 

 been digested with ammonia, is quite pure. 



The synthesis of chloride of silver was effected by heating 

 pure silver in a current of chlorine, and expelling the excess of 

 chlorine by dry air. In three experiments there were obtained 

 for 100 parts of silver, 132-841, 132-843, and 132-843 respect- 

 ively of chloride of silver. Chloride of silver was also prepared 

 by passing hydrochloric acid gas over the surface of a solution of 

 silver in acid. The precipitate was dried in the same vessel, and 

 fused in an atmosphere of hydrochloric acid. Two experiments 

 of this kind, one of which was made with 400 grammes of silver, 

 gave respectively 132-849 and 132-846 of chloride of silver from 

 100 parts of silver. In another series the silver was dissolved 

 in nitric acid, and precipitated by a feeble excess of hydrochloric 

 acid, washed, and fused in a current of the gas. This gave for 

 100 of silver 132-846 of chloride. In the last series, solution 

 of silver was precipitated by chloride of ammonium, the precipi- 



