356 Dr. James Walker on the 



osmotic pressure claims to put us in a position to calculate 

 the freezing-point of a dilute solution if we know its com- 

 position, the molecular weight of the substance in solution, 

 and certain easily determined physical constants of the sol- 

 vent — provided that the solution is not an electrolyte. If the 

 solution conducts electricity, then, in order to find its freezing- 

 point, we must in addition to the above data know its electrical 

 conductivity ; from which, by the help of the theory of elec- 

 trolytic dissociation, we may calculate the correcting factor 

 to be applied to the value deduced directly from osmotic 

 pressure. A specimen showing how such a calculation is 

 made will be given in the sequel. Mr. Pickering contends 

 that the results of experiment are not in harmony with the 

 values given by the theory ; and this is perfectly true as 

 regards the values calculated by Mr. Pickering, But then 

 he has repeatedly confounded the laws deduced from van 

 't Hoff's theory with the empirical relations stated by Raoult; 

 and, besides, has utterly ignored electrolytic dissociation in 

 his calculations. He says (Chem. News, lxiii. p. 171) : — 

 " Professor Arrhenius accuses me of neglecting the effect of 

 dissociation when discussing the freezing-points. In this he 

 is quite right, for I do not believe in it." Of course, the 

 assumption of such an attitude will undoubtedly enable Mr. 

 Pickering to find disagreement on comparing experiment 

 with calculation ; but, if the comparison is to be of any value, 

 it is surely evident that the theoretical numbers must be 

 legitimate deductions from the theory as expounded by its 

 originators, not as misconceived or distorted by its critics. 



Before proceeding further, we shall do well to look for a 

 moment at the conditions under which the law for the lowering 

 of the freezing-point is obtained from the hypothesis of 

 osmotic pressure. First of all, it must be understood that 

 the solutions considered are dilute; secondly, that they do 

 not conduct electricity ; and, lastly, it is assumed that when 

 a solution freezes, the pure solvent alone separates out in the 

 solid form. If these conditions are fulfilled, it follows from 

 the theory that the depression produced by one molecular 

 weight of a substance (in grams) dissolved in 1000 grams of 

 solvent, will be constant for any one solvent, and equal to 



*002T 2 



— ^ff— } where T is the freezing-point in the absolute scale, and 



W the heat of fusion of the solvent (van't HofF, Phil. Mag. xxvi. 

 p. 95). Further, it follows that the depression of the freezing- 

 point is proportional to the concentration of the solution (Blag- 

 den's Law). That this law is a well-justified generalization 

 where dilute non-electrolytic solutions are concerned, is amply 



