364 M. H. Rose on a new Series of Metallic Oxides. 



confirmed by a long series of experiments by Regnault, by which 

 he showed that the specific heat of the same substance may 

 be different according as it is taken in the solid, the liquid, or 

 in the gaseous condition — but that with simple bodies in the 

 solid form, which are otherwise in the same physical condi- 

 tion, the product of the specific heat into the atomic weight 

 varies between the numbers 38 and 41 (if the atomic weight is 

 taken at 100), and that these variations depend less on errors 

 in the experiments than on the fact that the specific heat of 

 bodies, as determined by experiment, contains foreign elements 

 which hitherto could not be eliminated — and that, further, 

 simple bodies in their various allotropic forms have a somewhat 

 different specific heat. 



Dulong and Petit found that the atomic weights of the then 

 known metals of easy preparation^ and of sulphur, were the same 

 as those which Berzelius had found, with the exception of sil- 

 ver. In the case of this metal, they found that the product of 

 the specific heat into the atomic weight, as assumed by Berze- 

 lius, only agrees with the numbers which the other simple bodies 

 give when the atomic weight is divided by 2, — a fact which 

 Regnault's experiments also confirmed. 



The second law which serves to determine the atomic weights 

 of simple bodies is that of isomorphism, discovered by Mitscher- 

 lich. According to this, compounds of the same atomic compo- 

 sition may, when they can be crystallized, assume the same forms. 

 That, exceptionally, there may be bodies of the same crystalline 

 form whose atomic composition could not hitherto be connected, 

 does not disprove the accuracy of this law. But both isomor- 

 phism, and the law of the connexion between the specific heat 

 and atomic weight of bodies, led to the conclusion that the atomic 

 weight assumed by Berzelius for silver must be diminished by 

 one-half. 



In an investigation of the crystallized compounds of antimony 

 and arsenic occurring in nature, I had found that the composi- 

 tion of the crystallized Fahlerz and Polybasites, which latter have 

 a simpler composition than the former, can only be explained by 

 assigning to sulphide of silver an analogous composition to sul- 

 phide of copper, Cu 2 S. Numerous analyses of these two mine- 

 rals gave the result that in them sulphide of silver and sulphide 

 of copper can replace one another in the most varied proportions*. 

 But the isomorphism of sulphide of silver and of sulphide of 

 copper, Cu 2 S, is still further confirmed by the composition of 

 crystallized argentiferous sulphuret of copper [Silberkupferglanz), 

 which consists of sulphide of silver and sulphide of copper, Cu 2 S, 



* PoggendorfFs Annalen, vol. xv. p. 576, and vol. xxviii. p. 156. 



