PALLADIUM. 95 



and 8.87 at 286°. They hence consider it probable that 

 the atomic weight of osmium is the same as that of plati- 

 num. The acid was very pure and was prepared by the 

 combustion of metallic osmium in oxygen. {Paris, Comptes 

 Rend., U. 1857, 1101.) 



OXYGEN. 



The atomic weight of oxygen is assumed at 16 for the 

 reasons stated under hydrogen, q. v. If hydrogen is taken 

 as unity, O = 15.96. 



PALLADIUM. 



Regnault determined the specific heat of palladium. It 

 corresponds to an atomic weight of about 106. {Gmelin- 

 Kraut, I. c.) 



J. J. Berzelius ; 106.61 (O = 16). 



In his earliest determinations of this constant, Berzelius 

 saturated the metal with sulphur, getting about 711 for S 

 = 201,165 ; and decomposed the chloride with mercury, 

 getting 704. [711 appears to be a misprint for 714.618 the 

 number given with corresponding data at Poggend., 8, 180.] 

 In this investigation a known weight of potassium chloro- 

 palladate was reduced in a current of hydrogen, the weight 

 of the residue determined, the potassium chloride leached 

 from the residue and the metallic palladium weighed. The 

 double salt was strongly heated, but not to fusion, in a current 

 of dry air before weighing. It being impossible to desiccate 

 this and the similar platinum-metal salts completely with- 

 out decomposition, the atomic weight was calculated irom the 

 relation between the metal and the KCl. Berzelius found 

 0.575 Pd = 0.809 KCl, and 0.851 Pd = 1.192 KCl. [If KCl 

 = 74.594 (Stas) the former gives Pd = 106.036, the latter 

 106.509.] Berzelius had reason to consider the latter analy- 

 sis the more accurate. {Poggend. Ann., 13, 1828, 454 / Kongl. 

 Vet. Acad. IlandL, 1828.) ' 



