CRITICAL EXPERIMENTS FOR DETERMINATION OF POTASSIUM, ETC. 623 



Experiment III. 



This experiment was carried on side by side of the preceding one, from 

 which it differed in this, that the rubidium was kept in excess over the platinum 

 solution, and that the latter was poured into the former. Platinum used 

 = 0*987 grm ; P = 132*6. Chloroplatinate dried at 130° C, but constancy of 

 weight not insisted on. Determination of total chlorine and of water again 

 omitted. Found per Cl 2 = 70 91 parts of chlorine ; M — 602*43 ; platinum 

 = 204*94; i.e., higher than ever before. [The synthesis gave 204*8 or 205*7 

 according to whether one or other of two limit values were adopted for the 

 uncertain titre of the rubidium solution.] 



Experiment IV. 



Here again an excess of rubidium was used, and the chloroplatinic acid 

 poured into the RbCl. Precipitate treated, and dried at 130° C, as before. 

 For its analysis, however, it was reduced by hydrogen in the dry way. 

 Platinum used — 0*991 grm. P = 133*8. Found per Cl 2 parts of fixed chlorine, 

 M = 611*08 ; platinum = 206*89. 



Experiment V. 



A parallel experiment to IV., from which it differed only in this that an 

 excess of platinum solution was started with, and the rubidium chloride poured 

 into it. Platinum used = 1*4861 grms.; P = 301. Found, per Cl 2 parts of 

 fixed chlorine, platinum = 198*19 ; M not determined exactly. Here again, as 

 in Experiment II., an excess of platinum used in the preparation of the chloro- 

 platinate prevented, to some extent, precipitation of surplus platinum. 



Experiment VI. 



In this experiment a very large excess of chloride of rubidium was employed ; 

 and the reduction of the chloroplatinate effected in the wet way. To make 

 sure of no chloroplatinate escaping reduction, it was divided into two parts 

 (each equal to about 1 grm.), each part dissolved completely in hot water, and 

 then submitted to the action of the hydrogen. To avoid loss of loose chlorine, 

 the hydrogen entered from the Kipp's apparatus, and passed out from the flasks 

 through U-tubes, the bends of which were closed with a layer of water. The 

 contents of these protection tubes, however, when tested after the experi- 

 ment with nitrate of silver, were found to contain no trace of hydrochloric 

 acid. 



Platinum used = 0*6917 grm.; P — 92*9. The chloroplatinate this time 



