Electric Residue in the Leyden Jar. 



485 



of the batteiy is great in comparison to that of the jar. It con- 

 sists simply in determining the quantity of electricity which will 

 be withdrawn from the battery by the jar. 



Before the beginning pf the experiment whose description we 

 have commenced, the sine-electrometer was connected with the bat- 

 tery of five jars, and the latter charged. After properly adjusting 

 the needle of the instrument it was moved a little backwards, i. e. 

 to an angle somewhat too small, and which amounted to 1 7° 53', 

 and allowed to remain, until, by the gradual loss of electricity in 

 the air, the image in the mirror coincided exactly with the mai'k. 

 At this moment the lever was allowed to fall by releasing a 

 hook i attached to a string ; a part of the electricity passed over 

 from the battery to the single jar, and the electrometer, which was 

 again adjusted with the least possible loss of time, in 15 seconds, 

 showed a deviation of 12° 13'. The square roots of the sines of 

 these angles are proportional to the charges Q and Q,' of the 

 battery before and after the charge was imparted to the single 

 jar. The latter charge amounts to Q — Q', and the proportion 

 between this and the quantity Q which was at first in the bat- 



Q — Q' 



terry, that is to say, the quotient — ^ — jWas calculated at 0*1695. 

 This method was often repeated, and gave the following results : — 



In Appendix II. the cause of the difference which still exists 

 between the several determinations of the quotient, amounting 

 almost to 5 per cent., will be further discussed. As, however, 

 the value of this quotient, according to the more trustworthy 

 method there adopted, is 01647, or very near the above, we 

 may set 



Q-Q' 



Q 



=0-165. 



Wc now know, therefore, that if at first the battery has a 

 charge Q, this will sink to 



Q' = Q(1 -0165) = 0-835 Q 

 at the moment when the single jar is charged. At the same 



