Properties of the Electric Field. 153 



metal in another ; and in neither of these cases could we 

 expect to get any change in the composition of the alloy at 

 the electrodes. We could only expect to find this when we 

 used an alloy in which the connexion between the constituents 

 could be regarded as of such a definite character that in the 

 molecule of the alloy one metal could be regarded as the 

 positive, the other as the negative element. The alloys used 

 by Prof. Roberts Austen do not seem to have been of this 

 character. The reasons which account for the absence of 

 change in the constitution of the alloy at the electrodes will 

 also account for the absence of polarization. 



Though the electrical conductivities of the metals are enor- 

 mously greater than those of electrolytes, there does not seem 

 to be any abrupt change from the conductivity in cases where 

 it is manifestly electrolytic, such as fused lead chloride, to 

 those in which it is not recognized as being of that nature, as 

 in carbon. The following table, giving the electric conduc- 

 tivity of some substances, will show this : — 



Silver 63 



Mercury 1 



Gas-carbon 1 x 10~ 3 



Tellurium .......* 4 x 1CT 4 



Fused lead chloride . . . . 2 x 10" 4 



There is a greater disproportion between the thermal con- 

 ductivities of silver and cement than there is between the 

 electrical conductivities of mercury and fused lead chloride ; 

 but no one argues that, on this account, the method by which 

 heat is propagated in silver is essentially different from that 

 by which it is propagated in cement. 



It is also suggestive that the substances which are inter- 

 mediate in their chemical properties between the metals and 

 the non-metals, such as phosphorus, selenium, and tellurium, 

 possess properties with regard to metallic conduction inter- 

 mediate between those of metals and electrolytes, — thus, as 

 shown by W. Siemens, the resistance of some of the modifi- 

 cations of selenium increases with the temperature, while that 

 of other modifications diminishes, — and that some of the modi- 

 fications seem to show polar effects,— -thus, when one electrode 

 is large and the other small, the current is greater when the 

 large electrode is negative than when it is positive. The 

 changes in the chemical properties of the substance seem to 

 proceed step by step with the changes in their behaviour with 

 regard to electrical conduction. 



If we accept the electromagnetic theory of light, we have 



