314 M. R. Kohh-ausch's Theory of the 



The cui-ves obtained from the observations in the tables a, b, c, 

 § 1, are therefore the actual represeutants of the disposable 

 charges L^ present at the diiferent times / in the apparatus, and 

 the residues p' sought in the tables a', b', c', § 2, denote also 

 quantities of electricity so circumstanced that « and a', b and b', 

 and c and c, are referred to the same unit respectively*. 



The first question which comes under our consideration is, 

 whether means are at hand which will enable us to obtain from 

 the observations a clear view as to the comparative quantities 

 of the residues formed in the charged jar. For the sinking 

 of the charge has, as we have observed, two causes ; the loss of 

 electricitv in the air, and the formation of the residue. If the 

 quantity of electricity Q^ be imparted to the interior coating of 

 the jar, which at the first moment shows the charge Lo=Qo, 

 then at the end of the time t the quantity Vf. is lost in the airt, 

 and a concealed residue rt is formed, so that the disposable 

 charge at present is only 



while the quantity of electricity Q< in the jar at the end of the 

 time t is expressed by the equation 



The observations gave us Qg and L^ directly, the latter at least 

 for the times / at which the disposable charge was observed ; and 

 the question now is, whether we can determine Vt so that the 

 residue for all observed times t may be given by the equation 



This leads us to the construction of a residue-curve, the nature 

 of which we shall be enabled to recognize ; and thus the hope is 



according to what has been said, eveiy distinct jar will have its own peculiar 

 unit. If, therefore, the sine-electrometer is to be made use of to compai-e 

 the chai-ges of chfferent jars, not the tensions on the knobs, but the quan- 

 tities of electricity actuallj- present, it is in the first place necessarj' to ascer- 

 tain the ratio of the quantities assumed as units in the two jars. This can 

 be found with certainty' by first causing the electrometer to show the same 

 divergence in both cases, and then chschai-ging the jars through the multi- 

 pljing galvanometer. 



* More was not necessary for our present pm-pose. The three different 

 units referred to in the tables a, b, c might indeed be reduced to each other. 

 Two other tables besides those to be used subsequently must for this pur- 

 pose be calculated, and the advantage did not appear to be worth this 

 trouble. 



t We intentionally avoid taking into account the loss of electricity by 

 the imperfect insulation of the edge of the jar, because otherwise we could 

 not set the loss as proportional to the charge Lt. The charge Q^ should 

 therefore be not taken greater than the vabdity of this assumption woi\ld 

 permit of. 



