3^6 Mr. brook’s Account of 
hollow on the fide next the wire H, fo as to fit, and is fc revved; 
in, lo as to prefs again ft H, ferves as a fpring to keep the ball 
F fteady, which Aides up and down, as well as turns round, 
on H. 
In order to make the divifions of Gi, fig. i. exactly a grain 
each, firft Aide the weight m towards the ball F, fig. i. till it 
is an exa£t counter-balance to the weight in F. At one end of 
the weight m let the divifions begin ; then fufpendanv tolerable 
pair of fcales, fo that the bottom of one of them may reft on 
the top of the ball r; then lay the ball Ii into the. fcale, and 
Aide the weight m near to Ii, and put as many grains into the 
other fcale as will juft raife the ball Ii in the fcale ; then mark 
the arm Gi at the fame end of the weight m y and divide the 
Apace between the two marks into as many parts as there are 
grains in the fcale, which may be divided and fub-divided into 
halves and quarters. 
The arm G2 being repelled ftiews when the charge is in- 
creafing, &c. ; and Ii tells what fuch a repuifive power is 
between two balls of the fize of thefe in grains, according to 
the number the weight m refts at when lifted up by the repul- 
live power of a charge. The longeft end of the index K ftiews 
how many degrees of a circle Gz is repelled'; and, by many 
trials, according to the number of grains, the arm Gi ftiews, 
when it is lifted up, and the weight m put at different places, 
fiuch reipeclive numbers of grains may be marked on the leaf! 
graduated circle on the dial plate where the Aiorteft end of the 
index points ; fo that when all the grains are thus marked on the 
dial plate, thus afeertained by the arm Gi, all thefe parts of the 
inftrument, that is, the ball F with the arms Gi and g may 
be taken off, and the inftrument is then graduated to be ufed 
without them ; butT do not know how the grains can be fo 
exaflly 
