VARIATION IN CONDUCTIVITY. 



63 



The best method, however, of comparing the variation in conductivity and 

 in fluidity is not by a direct discussion of the conductivity and fluidity curves ; 

 it is better to compare the respective ratios of decrease (by per cent) of the 

 two that is, the differences in per cent from what would be expected from 

 the rule of averages. These average values may be found graphically as 

 follows : At the extremities of a straight line of chosen length erect perpen- 

 diculars proportional in length to the respective conductivities (or fluidities) 

 of the two components of the mixtures, and connect these by a straight line. 

 Divide the base line into parts proportional to the composition of the various 

 mixtures. The perpendiculars joining these points of division and the line 

 previously drawn, will be proportional in length to the various average con- 

 ductivities (or fluidities). This can best be done by the use of coordinate 

 paper. The foregoing method has been used in making the comparisons 

 shown in table 39, from which it is seen that the variation in fluidity is in all 

 cases greater than the variation in conductivity. The two variations must 

 be compared in order to see where the effect of variation of fluidity is greatest. 



If the two effects were exactly equal, would equal zero. In this 



Aep 



relation we have a means of comparing the two effects. We have made the 

 comparison in table 40. 



TABLE 39. Variation (A^,,) in conductivity (percentage fall in conductivity) of 

 potassium iodide in mixtures of ethyl alcohol and water at 18. 1 



1 The conductivity data are taken from the work of Cohen (loc. cit.). 

 TABLE 40. Comparison of variations in conductivity and in fluidity. 



