1911-12.] Transverse Induction Changes in Demagnetised Iron. 313 
of the theoretical and experimental curves is obvious and may be traced in 
considerable detail. 
In fig. 14 (9 = 67 0, 5, the contribution made by each zone towards the 
transverse effect when H c is positive (fig. 1) is shown by the curves on 
the right ; when the rotation is towards — H c , by the curves on the left. 
Fig. 15 represents in the same way the results calculated for the com- 
plementary angle 6 = 22 0- 5. The curves on the left of both these figures 
cross each other, and in their initial stages finally pass from the third to 
the second quadrants for zones bounded by higher parallels, as they do 
in the corresponding experimental curves of figs. 11 and 12 when the 
residual magnetisation is increased. For higher zones (50° to 60° N. and 
others not here reproduced) the rise into the second quadrant is greater 
for the smaller (fig. 15) than for the larger (fig. 14) complementary angle. 
This phenomenon is well marked in the experimental curves. For reasons 
of symmetry, however, the positive portions of these curves must decrease 
as 0 is sufficiently reduced, and finally vanish with the transverse effects 
of which they form a part when 0 = 0°. 
But the final negative values of the transverse effects due to zones 
of north latitude (positive residual magnetisation) must vary as sin 0 . 
The positive portions of the curves do not therefore necessarily vanish 
when 0 = 90°, the angle at which the transverse effects reach their maxima. 
For reasons of symmetry, however, the differences between the curves on 
the right and left must decrease when 0 is sufficiently increased, each 
becoming the reflection of the other in the vertical axis when 0 = 90°. 
© 
Fig. 16 shows the curves calculated for the four north zones indicated 
when # = 90°. The curves are now symmetrical about the vertical axis; 
the equatorial zone typical of complete demagnetisation has vanished in 
the horizontal axis, above which all the other zones cross each other. 
The vertical rise is thrust towards the origin the further the zone is 
removed from the equator. 
These crossings were not obtained in the calculations for the curves on 
© 
the right of fig. 15 when 6 = 22 °’ 5 . They are indicated in the theoretical 
curves on the right of fig. 14 (the scale being too contracted to do more 
than this) when 6 = 6 7°*5, and these crossings appear to have their 
counterpart in the close spacing of the corresponding experimental curves 
of fig. 11 in their initial stages when the residual magnetisation is small. 
For this reason calculations were made for 0 = 90°, and finally these 
theoretical deductions were at a later stage experimentally verified in 
fig. 13, already shown on p. 310. No doubt the initial increase of the 
theoretical curves (fig. 16) indicates a much greater induction change than 
