ELECTRICAL PROPERTIES OF CONDUCTORS 593 



in comparison with the behaviour of pure iron, the character 

 of which will be gathered from the following table : 



TABLE V 

 Magnetic change in the resistance of iron 



Values of — x io 4 . 



J w 



From this it appears that the effect reverses between the 

 ordinary temperature and that of liquid hydrogen, and that 

 there is a neutral zone at about H = 7,000, where change of 

 temperature has no effect upon the resistance when under this 

 field. It should be noted that at this temperature, and even 

 more markedly below, a very small admixture causes an 

 enormous change in the electrical resistance, so that quite 

 small traces of impurity can be determined with great accuracy. 

 The effect is very marked in some other metals, for instance, an 

 addition of 2/ of gold to pure silver reduces the conductivity at 

 liquid hydrogen temperatures to about -V of its value. For the 

 consideration of the vibrators in the atoms and their relation to 

 the atoms themselves a study of the paramagnetism is instructive. 

 In a very large number of substances Curie found that the 

 susceptibility is inversely proportional to the absolute tempera- 

 ture. However, the majority of paramagnetic substances 

 deviate from this and approxmiate to one where the suscepti- 

 bility % = Or-- Also the deviations from Curie's law appear to 

 be governed by the law of corresponding states. With 

 platinum there is very little change of x with temperature as 

 also with many other elements. There is an interesting 

 difference between the behaviour of crystallised and anhydrous 



