Electric Residue in. the Ley den Jar. 311 



the electrometer are by no means proportional to the quantity of 

 electricity which was certainly present, but that, on the contrary, 

 a portion of this electricity became in some measure latent, that 

 is, was lost as far as the tension of the knob was concerned, and 

 we referred the effect to the concealed residue which was fornied. 

 It might appear rash to assume without further consideration, 

 that the portion which we saw returning as residue, r, after the 

 discharge of the jar, was, during the charging of the jar, altogether 

 without influence upon the tension of the electricity upon the 

 knob. If the residue, as is usual, be referred to the gradual 

 penetration of the glass to a certain depth by the electricity, 

 then it would seem most natural tq regard the total quantity of 

 electricity in a jar in which a residue is formed neither as pro- 

 portional to the tension on the knob of the jar, nor to the dis- 

 posable charge, but to something made up of both. At all 

 events we must, in the first place, ascertain how the matter 

 really stands. 



To decide this question, it is necessary to compare the same 

 tension, measured electroscopically, first of a batteiy where the 

 concealed residue is as small as possible, and secondly when it 

 contains a large residue, with the disposable charge, — the latter, 

 however, being measured in some other than an electroscopic 

 way. In such a case we may make use either of the thermic or 

 magnetic action of the cm-rent. The latter was the action 

 chosen. 



In the Appendix II. the small multiplying galvanometer is 

 described which served for this purpose, its magnetic needle, 

 furnished with a mirror and scale, being observed through a 

 telescope like a magnetometer ; and in the same place also the 

 arrangement is described by which the inner and outer coatings 

 of the battery, at the suitable moment determined by the sine- 

 electrometer, are connected with the galvanometer wire for a 

 time sufficient to allow the disposable charge to pass, but not 

 sufficient to permit any measurable portion of the residue to 

 make its appearance. 



When the current of a Leyden jar or battery, retarded by the 

 introduction of a liquid into the circuit, is sent through such a 

 galvanometer, and when the discharge, as in the case before us, 

 is accomplished in a small portion of a second, consequently in 

 a very small fraction of the time of oscillation of the needle, the 

 action of the current may be regarded as an instantaneous shock, 

 which imparts suddenlyto the needle a certain velocity of rota- 

 tion, C. This velocity is proportional to the strength of the 

 shock, and hence, in the present case, to the disjiosablc charge 

 sent through the galvanometer. Now W. AVeber, in his Electro- 

 dynamic Measurements, page 316, has given a formula, accord- 



