CHANGE OF ELECTRIC RESISTANCE OF NICKEL. 545 



aggregation. Anything which increases their velocities in a particular direction will 

 tend to increase the conductivity and to diminish the resistance. Anything which 

 tends to diminish the free path will increase the resistance. 



Consider the case of the nickel with a current passing through it. Here the 

 corpuscles will be driven against what is conventionally known as the direction of the 

 current. Let now a longitudinal magnetic field be applied. Its effect will be to give 

 the charged corpuscles a helical motion about the axis of the wire, and to drive them 

 out towards the outer surface of the wire. This will tend to increased aggregation of 

 the corpuscles with corresponding diminution of free path, and the result will be an 

 increase of resistance. Professor J. J. Thomson has shown that the manner in which 

 resistance increases with temperature in the case of metals must depend upon the 

 effect of change of temperature on the rate of dissociation of the molecules, and requires 

 the assumption that the effective collisions are those which occur between the negatively 

 charged corpuscles and the neutral molecules. In a recent paper by Mr J. Patterson 

 (Phil. Mag., June 1902) some calculations are made along the lines of this theory in 

 connection with experiments upon the change of resistance of certain non-magnetic 

 metals when placed in powerful magnetic fields. The problem is evidently quite a 

 different one in the case of the magnetic metals in which measurable changes of 

 resistance occur in low fields. 



Whatever be the ultimate structure of a molecule of iron or nickel it is almost 

 essential that it must be magnetic, the effect of an applied magnetic force being to 

 break up closed chains of molecules and give them a definite set more or less in the 

 direction of magnetisation. If we suppose these magnetic molecules to owe their 

 magnetic condition to whirls of electrified particles, it would appear that there was 

 greater chance for a corpuscle to be driven out of its whirl by a magnetic force acting 

 perpendicular to its axis than by a magnetic force acting parallel to its axis. Hence 

 it is quite conceivable that under a transverse force the resistance of the magnetic 

 metals should diminish, while under a longitudinal force it should (comparatively 

 speaking) greatly increase. 



In connection with this whole inquiry, an exceedingly interesting question will be, 

 How will the magnetic effect be affected when the temperature is raised so high that 

 the nickel ceases to be magnetic in the ordinary significance of the term ? This 

 question I hope soon to answer by means of an improved apparatus, capable of being 

 raised to very high temperatures. 



