Maynetic ileitis on Metallic Resistance. 905 



and tlii<, coupled with the transverse field where the wire 

 was bent back upon itself, undoubtedly lowers the maximum 

 and causes the slight drop in the curves. Curves IV. for 

 iron and III. for nickel show the effect of increasing this 

 transverse component. 



On the other hand, if the pole-pieces are set closer togethei 

 the lengths of the metal wires used must be correspondingly 

 decreased, and errors due to inaccurate setting are increased. 

 E. A. Owen * used wires about 1 millimetre long, and 

 obtained curves for nickel and iron in large magnetic fields. 

 For nickel he found the change of resistance reaching a 

 maximum at a field of 2800, and then decreasing steadily to 

 a minimum. For iron the maximum came at 1900; the curve 

 then turned downward and became a straight line, so that 

 ilR/R diminished as much as 18X10" 4 Erom a field of 1900 

 to one of 30,000. Clearly, Owen's results do not agree with 

 the curves obtained above for longer wires ; they are even 

 inconsistent with the more accurate measurements obtained 

 when the solenoid was used. Curve I. for iron continues 

 horizontal after passing Owen's maximum at 1900, and 

 curve I. for nickel shows no signs of a downward tendency 

 at 2200. The slight drop of III. (iron) and II. (nickel) is 

 much less pronounced than the slope of Owen's curves ; and 

 furthermore, this drop is a very evident result of a transverse 

 component. It seems probable that with very short wires 

 the field may be affected by polarity of the metal ; and this, 

 combined with the difficulty of setting such short wires 

 parallel to the field, will undoubtedly explain Owen's results. 



The conclusions arrived at may be stated as follows : — 



The increase of resistance of iron and nickel in longitudinal 

 magnetic fields approaches a maximum, but does not grow 

 less after reaching the maximum. For transverse magneti- 

 zation of these metals there is, as found by Grunmach, first 

 an increase and then a decrease of resistance as the magnetic 

 field increases. 



For cadmium the increase of resistance in both longitudinal 

 and transverse fields is, as demanded by Drude's theory, 

 approximately proportional to the square of the field strength. 

 The transverse effect is the larger. 



For the Heusler allov there is decrease in resistance in both 

 longitudinal and transverse fields ; and for large fields the 

 change is proportional to the field strength. In the case of 

 longitudinal magnetization, however, the effect may approach 

 a maximum, as measurements could not be made in larger 



* Phil: Mag. xxt. p. 122 (1911). 



