478 Dr. Harry C. Jones on the 



I. Into one univalent kation and one univalent anion, 

 II. Into two univalent kations and one bivalent anion. 

 III. Into one bivalent kation and two univalent anions. 

 IY. Into one bivalent kation and one bivalent anion. 



No new kind of ion formation from molecules will be found 

 among the remaining compounds with possibly the exception 

 of phosphoric acid, and they can all be referred to one of 

 the four classes. 



Magnesium sulphate dissociates much less than any com- 

 pound thus far examined. The decrease in the dissociation 

 with increase in concentration is very rapid. This decrease 

 from 0*001 n to O'Ol n is about 20 per cent., while from 

 0*01 n to 0"1 n about 28*4 per cent. This is why curve 8 

 (Plate IV.) rises so rapidly and approaches curve 7. 



Cadmium Chloride. — The cadmium salts were studied, 

 because the conductivity results showed an abnormally small 

 dissociation. These results also show that the chloride of 

 cadmium dissociates much less than the chloride of barium. 



Cadmium Bromide. — The bromide and iodide were mea- 

 sured that the effect of the three halogens combined with the 

 same kation might be seen. The bromide appears to have about 

 the same dissociation as the chloride in very dilute solutions, 

 but much less in more concentrated. At O'Ol n the difference 

 is but 2*3 per cent., while at 0*08 n this difference amounts to 

 16'2 per cent. 



Cadmium Iodide. — The iodide dissociates much less than 

 the bromide, and this difference is greater than between the 

 bromide and chloride. That the bromide occupies a position 

 between the chloride and iodide in reference to the amount 

 of dissociation was also found by Grotrian and Wershoven"*, 

 in their work on the " Conductivity " of these salts. The 

 curves 3, 4 (Plate V.) for the bromide, and 5, 6 (Plate V.) for 

 the iodide, are not as smooth as the curves for the other com- 

 pounds. The irregularities show some slight experimental 

 error, at some points increasing the lowering of the freezing- 

 point, at others diminishing it. The nature of this error with 

 these two salts is at present not apparent. 



Cadmium Nitrate. — The nitrate of cadmium dissociates to 

 a much greater degree than the chloride, and to nearly the 

 same extent as the alkali chlorides. 



Zinc Chloride. — The chlorides of barium and of zinc disso- 

 ciate to very nearly the same extent, while the chloride of 

 cadmium is much less dissociated than either. 



Hydrochloric Acid. — The solution of pure hydrochl oric acid 



* Wershoven, Zeits, pkys. Chem. v. p. 481. 



