KAriLENBERQ THEORY OP ELECTROLYTIC DISSOCIATION. 319 



unable to do this and have reached out to the hydrate theory 

 for help, it mighl be well in the face of the facts to acknowledge 

 freely the weakness of the dissociation theory and to proceed to 

 explain the facts on the basis of the hydrate theory alone. If 

 it be urged that the solutions conduct electricity and therefore 

 the salts must be electrolytically dissociated, — the answer is, 

 they must be, only if the dissociation theory be assumed. 



Attention must now be directed especially to the careful 

 determinations of C. Dieterici 1 , who measured the lowerings 

 of the vapor tensions of a series of aqueous solutions of elec- 

 trolytes and non-electrolytes at 0°. He found that for CaCl 2 

 the molecular lowering of the vapor tension diminishes strongly 

 with increase of the dilution, and then at about 0.1 normal 

 it again increases with the dilution. He also emphasizes the 

 fact that the very careful freezing-point work of Loomis 2 and 

 of Ponsot 3 shows a minimum of the molecular lowering of the 

 freezing-point at nearly the same concentration. The work of 

 Jones and Chambers above cited also corroborates this. Be- 

 tween the concentrations 0.1 and 1.0 normal, to which Dieter- 

 ici's work is really limited, he found that for the substances 

 tested (H 3 P0 4 , H 2 S0 4 , NaCl, CaCl 2 , cane sugar, dextrose 

 and urea) the molecular lowering of the vapor tension dimin- 

 ishes as the dilution increases, which is the opposite of that 

 which the theory of Arrhenius requires. In view of this fact 

 Dieterici rightfully refrains from even attempting to make 

 a further comparison between the degree of dissociation as 

 calculated from the vapor tension measurements on the one 

 hand and the conductivity on the other. The reader is also 

 referred to the able critical consideration of the very careful 

 freezing-point determinations of various investigators which is 

 contained in Dieterici's article. 



Attention is now called to the results of the boiling-point 

 determinations (Tables 12 to 29). For a number of the most 

 important salts represented in these tables, two, and in a few 



*1. c. 



^Wied. Ann. 51, 500 (1894); Ibid. 57, 495 (1896); Ibid. 60, 523 (1897). 

 •Recherches sur les congelations, Gauthiers and Villars, Paris (1896). 



