484 Proceedings of Roycd Society of Edinhurgli. [sess. 
proportion of bismuth ; in the bismuth-lead alloys also the variation 
decreased as lead was added — (compare fig. 4, diagram II., and fig. 
3, diagram I.). In the antimony-cadmium-bismuth alloy a small 
variation was observed. In the alloy of 6 parts antimony to 1 of 
zinc a small variation was also observed ; but in 806 antimony 
406 zinc a variation was not observed, nor could it be observed in 
the antimony-cadmium alloys. 
In all cases the variation was an increase. Only one position 
of the plate was considered, viz., that in which the plate’s length 
stood perpendicular to the direction of the magnetic field. 
The results would seem to indicate that the variation of resistances 
of an alloy in a steady magnetic field can he predicted from a Ttnoiv- 
ledge of the resistance variation of the metals composing the alloy. 
(c) The relation between the Transverse Effect and the Variation 
of Resistance. 
The equations used were 
and 
/\n = ±E 
+ ^2(7^)^ = ±V 
where A n and E are the variation of resistance and the numerical 
value of the transverse effect respectively as before defined. The 
relative values of the constants are given in the table of results ; 
the only peculiarity to be noticed is the negative value of Co for the 
alloy 19*5 bismuth 1 antimony; that is, that part of the transverse 
effect which is proportional to has the same sign as the 
component which is proportional to ( A n)^ ; the two effects working 
together give the extremely large negative result observed. In the 
other alloys of bismuth-antimony and in those of bismuth-lead the 
sign of C 2 is positive, and the result is that either the original 
negative effect becomes positive with higher fields, or at any rate 
reaches a maximum numerical value. 
