PLATINUM. 253 



If K = 39, and CI = 35.5, the first column gives Pt = 

 194.933. Seubert, from the percentage of platinum, gets 

 Pt = 194.392 ; and from the ratio 2KC1 : Pt he finds Pt = 

 194.494. 



As with the ammonium salt, three experiments were made 

 upon the potassium compound to determine the amount of 

 chlorine lost upon reduction in hydrogen. I cite the weigh- 

 ings, and add in a fourth column the quantity of KaPtClg 

 proportional to 100 parts of AgCl. This AgCl represents 

 hut four atoms of the chlorine: 



Sa/i'. Pi. AgCl. Ratio. 



6.7771 gnn. 2.7158 grm. 7-9725 gnn. 85.006 



3.5834 " 1.4372 " 4-2270 '■• 84.774 



4.4139 " 1.7713 " 5-2144 " 84.648 



Mean, 84.S09, d= .071 



Hence Pt = 195.002, ± .415. If K = 39, Ag =- 108, and 

 CI = 35.5, then Pt = 194.955. Seubert, calculating the per- 

 centage of chlorine and thence the ratio Clj : Pt, gets Pt = 

 194.631. 



Combining all the values we have the following result 

 for the atomic weight of platinum : 



1. From per cent. Pt in (NH^),,PtClg Pt = 194.314, dr .078 



2. " 6AgCl : (NH^)2PtCl6 ratio " = 196.871, ± .363 



3. '• percent. Pt in K2PtClg " = 194.370, rb .068 



4. " " KCl " " =194.645,^.213 



5. " 4AgCl : KjPtClg ratio " = 195.002, i .415 



General mean " =r 194.415. 4; .049 



Or, if O = 16, Pt = 194.867. 



Seubert, taking the arithmetical mean of his eight values, 

 gets Pt = 194.620. He regards, however, those results as 

 best which are dependent upon the percentage of platinum 

 in the ammonium salt, and upon the complete analysis of 

 the potassium compound. These give him a mean of Pt = 

 194.461, which, if corrected b}' reduction to a vacuum stand- 

 ard, becomes Pt = 191.34. 



In will be noticed that three of the ratios, calculated with 



