Magnetic Fields on Resistance. 817 



was experienced in eliminating contact resistance. The 

 usual method of copper-plating the ends of the rod and 

 soldering on connexions was used, and it was found that the 

 resistance of a given rod could be almost exactly duplicated 

 when the leads were taken off and new ones put on in the 

 same manner. Numerous observations were made on rods 

 cut in various directions from a single crystal, an octahedron. 

 A second crystal of the same form gave very similar 

 results. 



The curves obtained are especially interesting because they 

 show that magnetite, though crystallizing in a regular form, 

 has different properties along different axes. Magnetite T. 

 is a rod cut parallel to an axis joining two opposite vertices 

 of the crystal. Curve A is the longitudinal effect, both 

 current and field being parallel to the length of the rod. It will 

 be noticed that there is a slight decrease of resistance at first, 

 but for a field of about 200 the sign changes and the substance 

 behaves much like the ferromagnetic metals. The drop in 

 the curve after reaching a maximum is no more than can be 

 accounted for by inaccurate setting and the non-uniformity 

 of the field. 



Curve B is the transverse effect, the field being parallel to 

 an edge of the crystal. Here again there is a similarity to 

 the transverse effect in metals, especially is it like the curve 

 for the Heusler alloy. Curve is the transverse effect for a 

 field rotated 90° around the rod as an axis. 



Magnetite II. is a specimen cut parallel to an edge of the 

 crystal and perpendicular to I. Curve A is the longitudinal 

 effect and closely resembles the similar curve for Heusler 

 alloy. The tw T o transverse effects B and C, on the other hand, 

 are now very unlike. They show resemblance to the 

 longitudinal effect of I. 



Magnetite III. is cut parallel to an edge of the crystal and 

 perpendicular to I. and II. Curve A is for the longitudinal 

 field and B and C for the transverse. 



The relation between the curves can be best understood by 

 considering only the direction of H. In I. A, II. C, and 

 III. C, the field is in a constant direction with respect to the 

 crystal. These curves all have points of similarity. One 

 might expect I. A to degenerate into II. C or III. C if 

 the magnitude of the effect were made smaller, and setting 

 the current perpendicular to the field in II. C and III. C 

 has evidently produced this result. 



For the other curves such a relation is not easily seen. 

 II. A and III. B are respectively longitudinal and transverse 

 effects when H is parallel to one edge of the crystal, and 



