332 



MR. W. C. D. WHETHAM ON THE IONIZATION OF 



such as those due to self-induction or electrostatic capacity, are eliminated as soon 

 as a certain speed of the drum is attained, and beyond this speed variations in angular 

 velocity throughout a very wide range can be made without appreciably changing 

 the measured resistance. The only effect of such disturbances is to make the 

 galvanometer unsteady when the resistance of the solution falls below a certain value 

 (about 10 ohms in the apparatus used), and to make it rather less sensitive when a 

 very high resistance is measured, such as that of the water used as solvent. This 

 latter effect is chiefly due to the electrostatic capacity of the resistance coils as 

 usually wound, and it has been reduced by the employment of a new set of coils 

 specially wound in sections by Messrs. Elliott Brothers, as described by CHAPERON,* 

 which enables higher resistances to be accurately measured. 



As the drum revolves, very considerable thermo-electromotive forces are set up at 

 the contact of the brushes with the commutator, but these do not affect the galvano- 

 meter when the drum is in motion, for the currents due to them are reversed so 

 rapidly that they produce no resultant deflection. This can be proved as follows. 

 The galvanometer is disconnected from the circuit, and its position of equilibrium 

 observed. It is then connected up again, and, if the drum has lately been at work, 

 and the resistances in the circuit are low, a considerable deflection, due to the thermal 

 effects, is observed while the drum is at rest. The drum is then revolved, and the 

 galvanometer will be found to return to its normal equilibrium position. 



The success of the method of electrical measurement is best shown by the constancy 

 of the observed resistance of a solution when the current through it is changed, or an 

 alteration made in the ratio between the arms of the Wheatstone's bridge. As an 

 example the following numbers are taken from experiments on a barium chloride 

 solution : 



SECTION 5. On the Temperature to be used. 



The freezing point of a solution gives, as we have seen, a means of calculating the 

 ionization of the solute at that particular temperature. In order that the ionization 

 as calculated from the electrical data should be strictly comparable with this, the 

 ratio of the value of the equivalent conductivity of the given solution to that at 



* ' Comptes Rendus,' vol. 108, p. 799, 1899. 



