480 



SMITHSONIAN MISCELLANEOUS COLLECTIONS 



VOL. 54 



the concordance between liis determinations by very different methods 

 (a concordance verified by Halberstadt's investigation) precluded the 

 existence of errors due to impurities such as Dittmar and M'Arthur 

 assumed. 



The recent determinations by Archibald ' of the atomic weight of 

 platinum were based upon analyses of the platinchlorides and platin- 

 bromides of potassium and ammonium. In these analyses every precau- 

 tion was taken which modern experience had shown to be necessary. 

 The possible presence of moisture in the several salts was carefully 

 considered, and the potassium compounds in particular were dried at 

 380° to 400°. For the elaborate details of manipulation the original 

 memoir must be consulted. 



First, as to the analyses of potassium platinchloride. The salt, after 

 thorough drying and weighing, was reduced by heating in a stream of 

 pure hydrogen. The hydrochloric acid so formed was absorbed in water, 

 and afterwards converted into silver chloride and weighed. Ivnown 

 quantities of silver were used in this operation, so that two distinct 

 ratios were determined. From the residual mixture of ]3otassium chloride 

 and platinum the chloride was washed out, and its chlorine content was 

 estimated as in the previous determinations. The metallic platinum, 

 converted into sponge by again heating in hydrogen, was also weighed. 

 Vacuum weights are given in all of Archibald's determinations. The 

 weights were as follows : 



From these weights Archibald computes nine ratios as follows. In 

 the first ratio I have recalculated the figures into the percentage iorm 

 used for previous investigations. The other ratios are as Archibald gives 

 them; but with the probable errors computed by myself: 



1 Proc. Roj'. Soc. Edinburgh, 29, 721. 19f)9. 



