136 ; Messrs. Gee and Holden on 



of great influence in determining the stoppage of the current, 

 as is well shown by the following arrangement (see also 

 Table I.). Take two clean platinum electrodes, one being 

 much larger than the other, and place them in strong pure 

 H 2 S0 4 . It will be found that a very large current can pass 

 from the larger to the smaller electrode (see fig. 4, a direc- 

 tion of current), giving a deflection say of several hundred 

 divisions on the galvanometer. On reversing the current 

 (see fig. 5, /3 direction of current), if it be of suitable strength, 

 after the first sudden deflection the galvanometer-needle will 

 soon return nearly to zero ; in many cases scarcely a trace of 

 a deflection is to be observed ; in fact the current in the a 

 direction may be more than a thousand times as great as that 

 obtained in the /3 direction on reversing k ± . We desire to 

 ascertain the cause of this (almost total) stoppage of the 

 current, and to study generally the conditions determining 

 its occurrence ; and in connexion with this, it may be noted 

 that the formation of a sheath of gaseous bubbles round the 

 anode always attends this stoppage of the current with Pt 

 electrodes. 



b. Experiments proving Existence and Locality of an Insula- 

 ting Film. — If, after the production of the insulating condition, 

 we replace the kathode by a large electrode the current is 

 still stopped, but if the anode be similarly exchanged the 

 current passes readily. This experiment proves that there is 

 some obstructing cause at the anode. To determine its nature 

 attempts have been made to measure the polarization and re- 

 sistance of the cell in the a and |6 directions. Operating very 

 rapidly with the dead-beat galvanometer, and using Ohm's 

 method, we find that the resistance in the (3 direction (after 

 the insulating condition is produced) is very much greater 

 than in the a direction. Thus in one experiment the resist- 

 ance in the a direction was only 10 ohms, whereas on reversal 

 it rose to 10,000 ohms, and continued to increase for some 

 time after the insulating condition was produced. The pola- 

 rization, on the other hand, may be regarded as equal in the 

 two directions, for, using a compensation method in which the 

 voltameter-circuit was momentarily opened, and the polariza- 

 tion balanced against a known E.M.F., we have found 2*5 

 volts to be the value both when the insulating condition was 

 produced, and also when it was not. This result was con- 

 firmed by Ohm's method. Thus the polarization is not ab- 

 normal, and therefore cannot be a determining factor in pro- 

 ducing the insulating condition. Such a conclusion was 

 from the first to be expected, as the stoppage of the current 

 had been produced when a battery having an E.M.F. of ten 



