G2 
SIR WILLIAM THOMSON ON THE 
gives maximum effect in eacli case, and of the points in which the curves cut the line 
of abscissas, and hence of the relation between the magnetizing force of the earth and 
the Villari critical value. The magnetizing force for which 14 lbs. “ on” or “off” 
gave maximum effect was now found for the same wire to be about four or five times, 
and the Villari critical value about twenty-three times, the vertical component of 
the earth’s magnetic force at Glasgow. These results also disprove the negative 
maximum indicated by the curves of that notice, and show (as there stated in the 
Note at the end) that for higher magnetizing forces than the Villari critical value, 
the effect approaches a constant amount and the curves become asymptotic. 
210. The curves in Diagram III. were drawn by taking for ordinates the mean for 
each temperature of the effect for +M, and the effect for — M, of 14 lbs. “on” or 
“ off,” and for abscissas the corresponding current strengths, and therefore show 
approximately the effect which would have been produced by “on” or “off,” had the 
wire not been affected by the magnetizing force of the earth. 
By comparing the curve for cold water with the curve for hot water, we see that 
when the wire is at the temperature of 100° C., the average maximum effect of “on” or 
“off” is less than at the ordinary temperature of cold water by about 8 per cent, of 
the effect in the latter case, and that also, when the Villari critical value has been 
exceeded, the constant value to which the effect of “ on ” or “ off ” approaches is less 
for the higher temperature than for the lower, but in this case by about 30 per cent, 
of the amount of the effect for the lower temperature. The two curves also cross one 
another at a point above the line of abscissas, thus showing a greater critical value of the 
magnetizing force for the higher temperature than for the lower. 
211. The curves of Diagrams IV. and V. give at both temperatures the average for 
each strength of magnetizing current of the effects on -f-M and — M of applying and 
removing stresses of 7 lbs. and 21 lbs. respectively. 
A comparison of these curves with the average curves for 14 lbs. “on” and “off” 
(II. above) shows— 
(1.) That the effect at both temperatures of the application and removal of the 
stress is greater with 14 lbs. than with 7 lbs., and much greater with 21 lbs. than with 
14 lbs.; the maximums at the ordinary temperature in these three cases being respec¬ 
tively 31, 35, and 54. 
(2.) That the Villari critical value is much greater for 7 lbs. “on” and “off” than for 
14 lbs. : and, though by a smaller difference, greater for 21 lbs. than for 14 lbs. 
(3.) The difference between the maximum effects of “on” or “off” for the high 
and low temperatures is greater for 7 lbs. than for either 14 lbs. or 21 lbs., and seems 
to be greater for 21 lbs. than for 14 lbs. 
212. A series of observations of the effects of alternately making and breaking the 
circuit of the magnetizing coil and battery were made at both temperatures and for 
both +M and —M. The method of procedure was as follows— 
With no current flowing, a weight of 14 lbs. was placed in the scale-pan, and 
