64 ATOMIC WEIGHT DETERMINATIONS. 



that of silver at 107.928. [The sum of these weights is not 

 the molecular weight, and this, as well as recalculation of 

 the data, shows that the number is a misprint for 126.851. 

 Stas' results are, therefore, even closer to Marignac's than 

 his memoir would indicate.] 



Most of the experiments were made with iodine prepared 

 by the decomposition of nitric iodide decomposed in a large 

 volume of water at 65°. The iodine was further purified 

 by distillation over barium oxide and by other means. For 

 the preparation of silver see that metal. All possible pre- 

 cautions were observed in the preparation of all reagents 

 and in the conduct of the experiments. {Stas, Untersuch. 

 uber Chem. Prop. Leipzig, 1867.) 



IRIDIUM. 



Regnault determined the specific beat of iridium. It cor- 

 responds to an atomic weight of about 198. {Gmelin-Kraut^ 

 l.c.) 



J. J. Berzelius : 197.19 (0 = 16). 



Berzelius determined this value from analysis of potas- 

 sium chloro-iridiate. This salt reduced in hydrogen lost 29 

 per cent., the same quantity lost by the corresponding pla- 

 tinum salt, {vide platinum.) Berzelius originally calculated 

 the atomic weight of the platinum metals both from the 

 loss of chlorine of these double salts and from the relation 

 between the metal and the potassic chloride left after reduc- 

 tion. In his Lehrbuch he points out the impossibility of 

 complete desiccation, and resorts exclusively to the latter 

 method of calculation. With respect to iridium he merely 

 asserts that its atomic weight is the same as that of plati- 

 num, without there, or elsewhere, giving data as to the 

 amounts of iridium and potassic chloride found in the 

 reduced salt. It is, therefore, open to question whether he 

 assumed the identity from the loss on reduction or not. [It 

 Pt = Ir, and if KCl = 74.594, the value follows; see plati- 

 num.] Osmium and iridium were separated by fusion with 

 nitre, solution, and distillation. The residue was fused with 

 potassic chloride and sodium carbonate. On solution the 

 iridium remains behind. This residue was repeatedly roasted 

 and reduced to drive off osmium compounds. The potas- 

 sium chloro-iridiate was formed from the pure metal. {Pog- 

 gend. Ann., 13, 1828, 408; Kongl Vet. Acad. Handl, 1828.) 



