RICHARDS AND BONNET. — DISSOLVED CHROMIC SULPHATE. 23 



charge. The latter of these figures is improbably large, while the 

 former is plausible, approximatiug the values for zinc and magnesium. 

 Hence it seems probable that the number of charges does not greatly 

 exceed the number of atoms of chromium. It is not impossible that the 

 cation is CrO- or Cr(0H)2", which has been shown by Siewert (18G3) and 

 Whitney (1896) to be the most probable cation of boiled chromic chloride 

 and nitrate. On the other hand of course there is nothing to show that 

 the ion is not a complex one, with a valence about equal to the number 

 of chromium atoms, or at least a mixture of complexes with an average 

 valence no greater than this. This conclusion would be consistent 

 with that reached in other ways ; for if not all the normal valence of 

 chromium is available, it is obvious that some of it must be held by 

 undissociable hydroxyl, or hydroxyl and sulphuric acid, in a basic 

 complex. 



The absence of chromium from the portion migrating towards the 

 anode overthrows the hypothetical equation involving an ionized com- 

 plex acid, which otherwise seemed plausible.* 



Molecular Weights. 



The next step, suggested by the foregoing paragraph, is an investiga- 

 tion of the average molecular weights of the dissolved substances. In 

 mixtures so complicated the interpretation cannot be certain, but may 

 nevertheless afford some clue to the situation. It has already been 

 pointed out that the depression of the freezing point is the same for a 

 green solution hydrolyzed only by heating and a violet solution of equal 

 concentration, when sugar is present, — namely about 0.30° with a 

 twentieth molar solution. Since a molar solution should give a depres- 

 sion of 1.85° according to Raoult's law, there may be supposed to be 



20 X 0.30 



'- — = 3.2 active particles for each molecule of Cr2(SOj3 at this 



1.8o 



dilution, or the average molecular weight is 122. As has been already 

 suffCTested, this conclusion is somewhat in doubt, because of the presence 

 of su^ar, which may have an unknown specific action. Since much 

 ionized hydrogen is formed during the hydrolysis, it is obvious that the 

 remainder of chromium is probably condensed into a more complex com- 

 pound when the hydrolysis takes place — otherwise the number of active 

 particles would increase on hydrolysis instead of remaining constant. 

 This interpretation is supported by the freezing points of the more basic 



* Kriiger (1838) and Siewert (1845) ; also Richards, Zeit. anorg. Cliera., 23, 383. 



