226 Mr . Christie on magnetic 
nearly the same in the two cases, and yet in the sun the 
terminal arc is only 8°, whereas it is 12 0 30' in the shade. 
That the diminution of the arc does not arise from an increase 
of temperature in the box appears from this, that although 
the thermometer indicated 9 o° at the first observation in the 
sun, and 87° at the first in the shade, it is probable that the 
temperature of the brass compass-box was higher in the 
latter case than in the former, since in both this would change 
its temperature more slowly than the mercury in the bulb of 
the thermometer. 
To ascertain beyond doubt whether these conclusions were 
correct, I placed the compass in a room where it was not 
exposed to the rays of the sun, and when the needle had ac- 
quired the temperature of the room, I vibrated it, and noted 
the time of 40 vibrations and the arc at the 40th ; I then 
heated the compass-box over a fire until its heat was barely 
supportable to the hand, replaced and vibrated the needle. 
Instead of finding that the increased temperature of the box 
had diminished the terminal arc, I found that, at the end of 
the first 40 vibrations, it was considerably greater than when 
the box had been at a lower temperature, and that, as the 
box had cooled, the terminal arcs in the successive 40 vibra- 
tions decreased, the vibrations always commencing from 90° ; 
and this effect was invariably produced in repeated trials. A 
thermometer was placed at a small distance from the com- 
pass, that any changes in the temperature of the air might be 
indicated : this remained nearly at the same point throughout 
the observations. 
