On the Theory of the Electric Residue in the Leyden Jar. 413 



with the quantities of electricity Q< which were present in both 

 cases after 864. seconds. The bottle had a residue equal to 0"321 

 of this quantity, the plate 0-213. The whole hypothesis, how- 

 ever, is destroyed the moment we consider that according to the 

 above almost half as much electricity ought at this time to be 

 collected on the neck of the bottle, whose area is 11-4 square 

 centims., as on the 275 square centims. of the whole interior 

 coating ; a proportion which vnW be still more unreasonable if 

 we consider the state of the bottle at its final discharge after the 

 lapse of 5370 seconds. 



So much is clear, therefore, — the rim, if not entirely without, 

 has at any rate a very small influence on the residue. 



§7- 



In reference to this production of a residue, an influence has 

 also been ascribed to the cement with which the coating of tin- 

 foil is usually attached to the glass. Without entering into 

 theoretical considerations as to whether this influence must not 

 in every case be a vei-y secondary one, we shall here mention a 

 few facts merely. 



In the Franklin's plate c, the sine-electrometer was connected 

 with the metallic coating, which rests immediately on the glass 

 without the interposition of any cement. If we consider the 

 tables «", h", and c", and reduce the numbers of the two last to 

 the time 680, i. e. when the jar a, coated internally and exter- 

 nally with tinfoil, was discharged, the metallic coating certainly 

 appears to possess an advantage over this jar a, as well as over 

 the bottle b, filled with mercury ; for in 680 seconds the propor- 

 tion of the concealed residue to the charge imparted at the com- 

 mencement is 



a. b. c. 



0-297, 0-281, 0-187, 



and the proportion of the same to the total quantity Q^, which 

 was still present after 680 seconds, is 



a. b. c. 



0-324, 0-299, 0-204. 



According to this, the jar a, with the cement, furnished the 

 greatest residue, and that with the metallic coating the least. 



Apart from the fact that this diff'erence may be attributed to 

 the kinds of glass and to their thicknesses, the following experi- 

 ment is particularly worthy of notice. 



Before the bottle b, filled with mercury, was used for the 

 examination of the residue, it had been already filled with acidu- 

 lated water and examined, after having previously been washed 

 out with a solution of caustic potash, in order that the surface 



