249 



infiltration of electrical charge within the substance of coated and 

 charged electrics, and attributes it to a certain amount of conducting 

 power in the electrical substance. All substances, he infers, are con- 

 ductors of electricity in a greater or less degree, and thus admit of 

 infiltration of charge through their substance. In the case of charged 

 electrics, the infiltrated electricity subsequently returns upon its path, 

 and -hence residuary charge. 



The author has no disposition to question the experimental results 

 arrived at by either of these eminent men, but is of opinion that they 

 apply to a different case of electrical force than that of secondary and 

 immediate discharge, supervening upon a primary discharge of an 

 electrical jar, through an external explosive circuit, which he thinks 

 can neither be referred to any previous spreading of the charge upon 

 the glass, or to any penetration of it into its substance, or to return 

 action as described by Faraday. He has found that of 100 measures 

 of accumulated charge on a jar with imperfect conducting coatings, 

 no less than 75 measures, or three-fourths nearly of the whole accu- 

 mulation, has been left behind after explosive discharge. In a jar 

 coated with water, full 14 measures out of 100, or about one- 

 seventh, was left undischarged. He thinks it difficult to reconcile 

 such an amount of residuary charge as this, with any spreading of 

 the electricity on the glass, or any possible amount of penetration 

 into its substance. 



Although the deductions of Cavendish and Faraday may not be 

 found to apply as solutions of the interesting problem of residuary 

 charge, they still find their application in other cases, as in the case 

 of the facts noticed by Nicholson already detailed. The intensity of 

 explosive discharge may apparently become increased by a penetra- 

 tion of the exploding electricity into the air separating the balls of 

 the discharging electrometer, in which case the measured distances 

 of discharge, according to Cavendish, would, for given measured 

 quantities of electricity, continually decrease, and discharge at the 

 measured distances between the exploding balls would appear to 

 happen prematurely. It is now shown by reference to a Table of 

 experimental results, that at distances 1, 2, 3, 4, taken in tenths of 

 an inch, with quantities of measured charge also as 1,2, 3, 4, the 

 actual distances of explosion are nearly as *1, '214, '325, *445. The 

 author hence infers that, supposing the penetration of the first 



