ELECTRICAL ACTIVITY OF ORE BODIES. 



863 



etc. The values of e, as derived both from the direct and return series, are 

 given in the table, the latter being primed. Heavy black lines across the 

 table indicate either that the bags were refilled or that the experiments had 

 to be temporarily discontinued. 



Table XXL, containing £ x 10^ 



The successive values of e are not constant, though in the majority of 

 cases they are so small as to be immaterial. At times, however, values suffi- 

 ciently large to be important are reached. An exchange of terminals is 

 therefore indispensable, especially as experiments will usually be sufficiently 

 extensive to involve interruptions. The gradual variation observed in the 

 value of £ can most probably be referred to a corresponding change in the 

 concentration, etc., of the solution (zinc sulphate) contained in the bags. It 

 is hardly probable that it is due to polarization, or a change in the surface of 

 the amalgamated zinc strips. It is interesting that, in spite of the fact that 

 for the holes I., II., and III. the electromotive force between the bags in the 

 direct and return series differs largely, the lode-currents deduced from the 

 two sets of data are practically equal. (See Table XV.) 



Wire. — In the above experiments especial care was taken to prevent 

 errors due to leaks in the wire. The galvanometer was sufficiently delicate 

 to register a fault of this kind of 6,000,000 ohms' resistance with certainty. 

 As has been mentioned, every leak introduces an electromotive force 

 zinc [copper;* hence the great necessity, notwithstanding the fact that the 

 latter must act through a very great resistance, of avoiding leakage. 



'For we have the closed couple: Copper (of wire); liquid (moist earth, etc.); zinc (of bag). 



