332 Mr. L. Lownds on the Thermomagnetic and 



Case 2, § 3. The chief crystallographic axis is at right 

 angles to the current of heat, and the magnetic lines of force 

 perpendicular to the plane of the plate. The results for 

 mean temperatures 53°'7, 4°-4, -41°"9, and -94°'7 are 

 given in Table II., and plotted as the broken -line curves in 

 fig. 5. 



It will be noticed from the tables that there is a dissym- 

 metry for the two directions of the field which is more 

 marked in the case in which the current of heat runs at right 

 angles to the chief axis. The values for the two directions of 

 the field vary with the setting up of the crystal, although the 

 mean values are of the same order. The dissymmetry is 

 probably due to the current of heat not flowing exactly in 

 the direction indicated. Such a deviation would introduce a 

 transversal component which would change its sign with the 

 field, and would therefore disappear on taking the mean of 

 the two values. 



Fi<r. 5. 



<% 



•>*. 



20 



LO/\C/rUl)/NAL 



EFFECT 



s ^ 



yystt 



Vo'qrc^ ihic a xis 



±* 



JO 



40 



From an examination of fig. 5 one sees that : — 



1. The curves connecting field-strength and longitudinal 

 effect are of a similar form to those for electrolytic 

 bismuth, but the negative values are obtained at 

 much higher temperatures. 



