ON THE HALOGEN HYDRIDES AS CONDUCTING SOLVENTS. 155 



In order to test the conclusions that have been arrived at, we require to know the 

 concentration of the unassociated solute for each dilution. 



If A gram-molecules of the solute are dissolved in one litre of solvent, and if 

 there are formed c gram-molecules of the electrolytic compound, then, if n molecules 

 are required to form one molecule of the compound, nc molecules of solute will have 

 been used, when equilibrium is established, so that the equilibrium equation is 



k(A-nc) a = k'c, 

 or 



c = K (A-wc)", 



and the general expression for the specific conductivity becomes 



ic = aK'(A-wc)". 



No solution of any substance in any one of the halogen hydrides lias yet been 

 found with high value for the conductivity, a fact which may be assigned to one of 

 three causes, namely, either (1) the concentration, or (2) the coefficient of ionisation of 

 the electrolyte is small, or (3) the ionic velocities may be very small. If we assume 

 the first to be the most probable cause, A nc will not differ much from A, and we 

 may without sensible error make use of the values for the total concentration in 

 applying the above equation to our results. 



This has been done, the equation being used in the form of /cV" = aK', and the 

 results of the calculations are given in Tables I. and II. 



Table I. contains the values of V and of /cV" (or aK 7 ) for those solutions in which 

 n = 2, that is, in which two molecules of solute combine with the solvent to form 

 one molecule of the electrolyte. Table II. contains the similar value for those 

 solutions for which n = 3. 



It will be noticed that in some cases the figures exhibit considerable irregular 

 variation. This is to be expected from the fact that the measurements of conductivity 

 and of concentration are subject to considerable experimental error. These errors 

 were not specially guarded against, as our object was to establish beyond question 

 the nature of the variation of p. with V rather than to obtain accurate measurements, 

 which, in the present state of our knowledge, would not possess any special value. 



The figures for K at very high dilution are, in some cases, quite valueless as a 

 test of our hypothesis, on account of the enormous influence of very slight errors of 

 observation at these dilutions. 



The results contained in Tables I. and II. are shown graphically in figs. 1 and 2 

 respectively. 



x 2 



