Analysis by Spectrum-observations. 503 



and deliquesces. When ignited for a long time in contact with 

 air it becomes slightly basic. 



According to the atomic-weight determinations, already de* 

 scribed, 1-0124 grm. of chloride of csesium, the solution of 

 which was perfectly neutral, yielded 0*9133 grm. chloride of 

 silver, and - 0009 grm. of metallic silver from the filter-ash. 

 This corresponds to the following numbers :— 



Calculated. Found. 



Cs . . . . 123-35. 77-67 77'67 

 CI ... . 35-46 2233 22-33 



158-81 100-00 100-00 



i. Double Chloride of Platinum and Casium. 



If, to an aqueous solution of chloride of caesium, bichloride of 

 platinum be added, a yellow precipitate is formed. The colour 

 of this is somewhat lighter than that of the corresponding potas- 

 sium-salt, because it is less soluble than the latter, and therefore 

 is deposited in a finer state of division. The precipitate is anhy- 

 drous, and is composed of microscopic, honey-yellow, transparent 

 regular octahedrons. 100 parts of water dissolve of this com- 

 pound — 



at C. 0-021 part. 

 11 0-072 „ 

 40 0-118 „ 



at 68 C. 0-234 part. 

 100 0-382 „ 



These numbers are taken from the mean of a large number of 

 careful determinations agreeing well amongst themselves. 



As almost all the platinum which is found in'commerce is very 

 impure, and often possesses an atomic weight from 6 to 8 per cent, 

 below the true value, we have previously purified the platinum 

 which we used for the preparation of these as well as of the rubi- 

 dium salts. This purification was effected by fusing the chloride 

 of platinum and potassium in a platinum dish with a mixture of 

 the carbonates of potassium and sodium, washing out the mass 

 with water, and dissolving the residue in dilute aqua regia. When 

 this operation had been repeated five times, it was found that the 

 platinum attained an atomic weight varying but very slightly 

 from 99-1. 



The analysis of the double chloride was carried out as fol- 

 lows : — The salt was weighed out in a U-shaped tube of hard 

 glass, after having been dried in a bath of fused chloride of zinc 

 at a temperature of 160° to 170° C, and the tube, with the sub- 

 stance, bedded in magnesia and heated to dull redness, whilst a 

 current of dry hydrogen was passed over the salt. The loss of 

 weight thus obtained was determined, the chloride of csesiuni 



