RECENT PROGRESS IN PHYSICS. 375 



handles, the two electrified places are touched simultaneously, or so rap- 

 idly one after the other that the contact may he considered as simul- 

 taneous. One of the proof halls is then placed in a large hell glass, 

 the other applied to the torsion balance. After measuring the torsion 

 for the first proof hall, it is removed and the second, (kept meanwhile 

 in the bell glass,) is applied to the balance, and the corresponding 

 torsion, (with equal elongation,) is measured. The times at which 

 the two readings are made, are ol)served by means of a watch marking 

 seconds. The torsion is now diminished a few degrees, hence the 

 elongation is somewhat increased, and the length of time the balance 

 beam occupies in returning to its former position is then noted. 



An example may serve better to explain this metliod of observing. 

 Suppose the two proof balls I and II have been applied to the places 

 whose electrical densities are to be compared. 



I being brought to the torsion balance requires a torsion of 55°. 5, 

 to bring' the beam into a certain position, that is, deflected a certain 

 number of degrees. 



II is now applied. To bring the beam into exactly the same posi- 

 tion as before, the thread must receive a torsion of 293°. 5. 



Between the first and second reading a period of 3.1 minutes has 

 elapsed. 



Now suppose the torsion is reduced 20° or to 273°. 5, and, counting 

 from the second reading, 3.2 minutes elapse until the beam returns 

 to its former position. 



We will now proceed to the computation of these data. In 3.2 

 minutes proof ball No. II has lost a quantity of electricity which is 

 measured by a torsion of 20° ; if the loss of electricity is proportional 

 to the time', the loss of No. II between the first and second reading 

 amounts to — 



3.1 



— X 20 = 19°. 4; 

 3.2 

 but the loss of electricity is proportional not only to the time, but also 

 to the density, which is not equal in the two periods.^ We may sup- 

 pose, without sensible error, that the electrical density of proof ball 

 II, in the first period, is to the density in the second period as 293.5 

 is to 273.5; hence the loss of ball 11 in the first period, (that is, 

 between the first and second observations.) would be — 



293.5 



19.4 X =20°. 8. 



273.5 

 At the instant in which the electrical charge of ball I was measured 

 in the balance, the charge of ball [I was 



293.5 + 20.8 = 314°. 3, 

 Hence, the required proportion between the two quantities of electri- 

 city is 



If we indicate by a the torsion first measured, by I the second, by c 



