AND ITS APPLICATION TO TORSION BALANCES. 
221 
the number of degrees of torsion which the thread has suffered. Remove the 
weight, and untwist the thread till the beam returns to its horizontal position. 
Put a small known weight into the same scale, and turn the torsion key till 
the beam be raised to its former horizontal position, and observe the number 
of degrees of torsion : — then will the degrees of torsion give the ratio of the 
known and unknown weights. For example, if the thread suffered a tor- 
sion of 1500 degrees to raise the body B, and 1000 to raise one grain, then 
1000 : 1500 : : 1 grain : 1.5, the weight of the body. If the body required only 
50 degrees of torsion to raise it, then its weight would be ^ or .05 of a grain. 
2ndly. When the body to be weighed is much heavier than a grain, the best 
way will be to ascertain its weight within a grain by the method of double 
weighing, and then apply the principle of torsion for ascertaining the fraction 
of a grain. Suppose the body to weigh nearly 100 grains, twist the thread 
through two or three circumferences exactly as in the former method. Bring 
the beam to a horizontal position by small shot or filings. Put the body into 
one of the scales, and shot or filings into the other, till the body be exactly 
counterpoised. Remove it, and substitute known weights till they be nearly 
equal to the weight of the body. Turn round the torsion key till the beam be 
brought to a horizontal position, and note the degree of torsion. 
Put a grain into the scale, and observe the additional number of degrees of 
torsion necessary to bring the beam to its horizontal position ; and we thus 
get the fraction of a grain which the body exceeds the known weights em- 
ployed. If for example the body weighed nearly 100 grains, and it required a 
torsion of 50 degrees to raise the scale when 99 grains had been put into it 
after the body had been removed, and also that the thread required an additi- 
onal torsion of 1000 when one grain had been placed in the same scale, then 
the weight of the body will be 99^ or 99.05 grains. 
8. It is of course necessary to prevent the agitation of the air acting on 
the balance and its scales, and therefore the whole may be inclosed in a box 
as in the common balance. It is not necessary, however, to inclose the glass 
thread and the divided circle, and therefore the thread may be made to pass 
through a hole in the back of the box, and removed when the balance is not in 
use. It is useful to have a number of threads of glass of different degrees of 
