Induction of instantaneous Electric Currents. 499 



rounded the cylinder, the re-electrometric deviations were no 

 longer wanting. 



I suspended the connexion between the little copper cylin- 

 der and the wire of the instrument, and having connected it 

 instead with the ends of the coil which surrounded it, and then 

 discharged the Leyden jars upon the cylinder itself, the usual 

 deviations no longer failed. Are not these phaenomena de- 

 pendent uponLeyden-electrical actuation ? As I invariably saw 

 that, when I made the discharges pass through two points of 

 the coil itself, there were very nearly the same deviations ; and 

 in the other case (namely, when the copper cylinder was con- 

 nected with the wire of the re-electrometer) there was exactly 

 the same effect when I discharged the jar connecting it with 

 the two ends of the cylinder, I doubted whether, instead of 

 inductions, I might not hitherto have seen only the effects of 

 a division and subdivision of the discharge. 



. V. It appeared to me that I had taken the most scrupulous 

 precautions that there might be no metallic contact between 

 the actuating wire and that to be actuated ; but might it not 

 be that even through the silk the electricity might make itself 

 a passage ? The doubt was so much the more reasonable, as 

 from the experiments related in the preceding memoir I had 

 learned that the currents of the Leyden jars might divide 

 themselves between good and bad conductors ; I sought there- 

 fore to clear up this doubt by the following experiments. 



I put the ends of the wire itself in metallic communication 

 by means of a band of lead two centimeters broad and eight 

 decimeters long; I covered one part with a small piece of very 

 dry wool ; then upon this wool and above the band itself I ex- 

 tended for the space of a decimeter, another band of the same 

 metal, which rose at a right angle from both parts, for the 

 space of four centimeters. Having many times discharged 

 the Leyden jar, now in one manner, now in the other, putting 

 the coatings in contact with the extremities of the second 

 band, deviations of two, three, and even six degrees in the 

 needle of the re-electrometer were always obtained. 



I repeated the same experiments after having put four small 

 pieces of wool between the band of lead adjoining the ends of 

 the re-electrometric coil and that upon which the jars were 

 discharged, and the results were pretty nearly the same. * 



The deviations were somewhat smaller (though they never 

 failed) when, besides the said pieces of wool, I also placed 

 between the two metallic bands a large cake of sealing-wax 

 having a strongly insulating power, and six good millimeters 

 in size. And as in these experiments I made use of a small 

 jar highly charged, and the spark passed thence to a great 



2L2 



