274 Wisconsin Academy of Sciences , Arts and Letters. 
vibration. For needles 34 lines long they found the following formula 
held: 
I' = I [1 - 0.000324 (t' -1) d], 
where I and I' are the intensities of magnetization at the temperatures 
t and t' on the Reamur scale, and d is the diameter of the magnets. For 
needles 2 inches long the numerical factor is 0.000432, showing that the 
proportional change in intensity of magnetization is greater in shorter 
magnets. Their temperature limits were 0° and 80° R. By swinging 
their magnets at different temperatures they found the change in moment 
proportional to the difference in temperature as is shown by their for¬ 
mula, which is to be applied with caution for their magnets had not 
been brought to what we call the permanent state. 
In 1851,* Lamont found that when a permanent magnet was alter¬ 
nately heated and cooled fifteen or sixteen times between fixed limits 
of temperature, it reached a permanent state in which it had a definite 
magnetic moment for a given temperature and always returned to that 
moment when brought to the corresponding temperature, provided only 
it had never passed beyond the temperature limits mentioned above. 
The higher the temperature, the smaller was the magnetic moment. 
tProf. G. Wiedemann has made some careful investigations on the 
influence exerted by the temper of the steel and the original intensity 
of magnetization. He used bars 22 cm. long and 1.35 cm. in diameter. 
Before being magnetized, they were alternately placed in melting snow 
and boiling water fifteen times, in order to bring the steel itself as far as 
possible into such a state that alterations in temperature would produce 
no structural change. The bars were magnetized in a coil at a tempera¬ 
ture of 0°. They were then carefully placed in a box of sheet copper be¬ 
fore the needle of a magnetometer and the deflection was observed by 
a telescope and scale. The temperatures of 0° and 100° C. were obtained 
by means of melting snow and boiling water. His results for magnets 
that have reached the permanent state show that in the case of hard 
steel magnets the change in moment is nearly proportional to the 
moment at 0°, while for tempered and soft steel magnets, the ratio of 
the change to the moment at 0° C. increases with the moment. As his 
results give a good idea of the size of the changes under discussion, I ap¬ 
pend the following table from his paper: 
* Lamont, Pogg. Am. 82, p. 440, 1851. 
+ G. Wiedemann, Pogg. Am. 100, p. 235, 1852; 103, p. 563, 1858; 122, p. 355, 1664. 
