SUPERPOSED MAGNETISMS IN IRON AND NICKEL. 



197 



Group D. — Cross Section of Iron Wire = "000714 sq. cc. 

 Experiment I. Linear Current = 0'65 Arnp. 



Helical Current, 



0-508 



0-995 



1-593 



2-262 



2-615 



3-157 



4-068 



Tension. 



( 



103 



290 



368 



348 



336 



329 



265 



388 



Twist, . .< 



142 



400 



613 



619 



613 



574 



484 



258 



1 



129 



484 



697 



761 



787 



761 



723 



129 



Experiment II. Linear Current = 0-973 Amp. 



Helical Current, 



0-508 



0-995 



1-593 



2-262 



2-615 



3-157 



4-068 



Tension. 



( 





... 



794 



839 



865 



807 



774 



388 



Twist, . .< 



... 



... 



923 



1090 



1077 



1013 



884 



258 



( 



252 



613 



916 



1045 



1123 



1084 



1045 



129 



The direction of twist was as found by Wiedemann. If the current is 

 passed doAvn the wire from the fixed to the free end, and the wire is mag- 

 netised with north pole downwards, the free end, as looked at from above, 

 twists in the direction of the hands of a watch. As pointed out by Maxwell 

 and Chrystal, this agrees with Joule's discovery mentioned above. For the 

 circular magnetisation due to the down current is right handed with reference 

 to the current. Hence the resultant magnetisation lies in a direction inter- 

 mediate to the circular and longitudinal magnetisations at any point ; and as 

 the iron extends in the direction of magnetisation, and contracts at right angles 

 thereto, there will be a lengthening of the wire in a direction oblique to the 

 axis, such as to cause a twist in the direction specified. The amount of twist 

 depends not only on the magnetising force in this oblique direction, but also 

 upon the obliquity, so that a maximum twist for an intermediate value of the 

 helical current is quite in accordance with Joule's result that the extension 

 increases with the magnetisation. Suppose, for example, that the circular and 

 longitudinal magnetisations at a point on the wire are a and fi, and that these 

 give a resultant magnetisation */a 2 + /3 2 in a direction making an angle, whose 

 tangent is a//3, with the vertical line through the point. Let the extension 

 along this direction be represented by/x (a 2 +/3 2 ), an assumption approximately 



