PLATINUM. 251 



Latest of all we have to consider the experiments of 

 Seubert.* This chemist prepared very pure chloroplati- 

 nates of ammonium and potassium, and from their compo- 

 sition deduced the atomic weight of the metal under con- 

 sideration. The ammonium salt, (NH ^ ) ^ PtCl g was analyzed 

 by heating in a stream of hydrogen, expelling the excess of 

 that gas by a current of carbon dioxide, and weighing the 

 residual metal. In three experiments the hydrochloric acid 

 formed during such a reduction was collected in an absorp- 

 tion apparatus, and estimated by precipitation as silver 

 chloride. Three series of results are given for the percent- 

 age of platinum in this salt, together with another single re- 

 sult which may be considered alone. Here are the figures : 



Series I. Series IT. Series III. 



43-957 43-871 43-990 



43.948 43-876 43-986 



43.960 43-872 44.001 



" 43-946 43-881 44.020 



43-963 43-875 43-994 



43-961 43-879 43-996 



44.004 



Mean, 43.956, ± .002 Mean, 43.876, dr .001 44.026 



43-998 



Mean, 44.001, zb .003 



These series represent three preparations. The additional 

 single experiment above referred to was made with material 

 belonging to series II, but recrystallized from water. This 

 salt gave 43.955 per cent, of platiimm, a figure to which we 

 may assign the probable error of one experiment in the 

 first series. Combining, we get the subjoined general mean 

 percentage of Pt in (NHJ^PtClg : 



Series I 43.956, zb .002 



" II 43.876, dz .001 



" III 44.001, ± .003 



Extra experiment 43-955> ± -004 



General mean 43.907, dz .0009 



*Ber. der Deutsch. Chem. Gesell., 14, 865. 1881. 



