116 ACTION OF AQUEOUS AMMONIA ON 



resulting mixture should be 2(200-|-270.9) = 235.5, or the same 

 as found by experiment. This possibility led to many experi- 

 ments designed to prove the presence of free mercury in the 

 vapor^. Though these experiments indicated that some dis- 

 sociation took place, for the most part they failed to give any 

 idea of the extent. A. Smith and A, W. C. Menzies^, how- 

 ever, concluded from their work on the vapor pressure of calomel 

 that neither HgCl nor HgjClz was present in the vapor. The 

 case seems to be analogous to that of ammonium chloride. 

 Ordinarily, ammonium chloride dissociates on volatilization to 

 give a gaseous mixture of ammonia and hydrog-en chloride. 

 But if the ammonium chloride be dried with extreme care before 

 heating, then the dissociation does not take place. H. B. 

 Baker"* has obtained results with very dry mercurous chloride 

 indicating that the molecules in this condition exist in the gaseous 

 form chiefly as HgaCU but dissociate in the presence of a trace 

 of moisture to Hg and HgClj. 



A. Smith and A. W. C. Menzies^ found that mercurous 

 chloride, dissolved in mercury, lowered the vapor pressure of 

 the mercury to an extent equal to the lowering calculated on 

 the assumption that the molecule of calomel is HgCl. On the 

 other hand, E. Beckmann^ obtained the molecular weight 

 corresponding to HgaClj by the freezing point method, using 

 either mercuric chloride or anthraquinone as the solvent. 

 Similarly, he obtained the formulae HgjBrj and Hg2l2 by noting 

 the effect of these halides on the freezing points of the corre- 

 sponding mercuric halides. 



From the point of view of the modern theories of atomic 

 structure, it is hard to conceive of the molecule HgCl existing 

 to any extent at ordinary temperatures. According to C. R. 

 Bury^, the electron structure for mercury is 2.8.18.32.18.2. 

 It is easy to see that an atom of such a structure would lose two 

 electrons readily to chlorine for instance. But to lose only 

 one electron ought to leave a very unstable and extremely 

 reactive substance. The case is in no wise analogous to that 

 of copper. Here, according to the same hypothesis, the two 



