Intelligence and Miscellaneous Articles. 235 



of the needle, the length of which was normal to the radius of the 

 disk, was such as would have been produced by a positive electric 

 current flowing with the rotation of the positively charged disk, or 

 against the rotation of the disk charged negatively. 



There was no alteration in the action when the gilding of the 

 ebonite plate was removed in a series of radial lines, so that annu- 

 lar electric currents could no longer take place. A thin plate of 

 glass was also inserted instead of the gilt ebonite, and, like the 

 disk of a Holtz machine, could be electrified through points ; while 

 close beneath it there was a gilt resting plate connected to earth, 

 in order to fix as much electricity as possible. The direction of the 

 deflections was the same as in the previous experiments ; but they 

 were smaller, as the conditions were not so favourable for strong 

 electrification. 



In order to compare the electricity carried forward by convection 

 with that which passes in conductors, experiments were instituted 

 in the following manner. 



The ebonite disk was gilded afresh, and the gold coating divided, 

 by a series of fine circular lines, into rings insulated from one 

 another. The innermost ring was connected with the axis ; the 

 rest could not at any rate become considerably charged without 

 discharging themselves by very short sparks from one to the other. 

 Two electrified plates, each having the form of a sector of a 

 circle, but which did not reach to the axis, were placed, opposite to 

 one another, above and below the rotating plate. Under these 

 circumstances the electricity of the gold rings must have been ac- 

 cumulated by electrostatic induction in the sector covered by the 

 last-mentioned plates, and carried forward convectively. TThen 

 this electricity was positive, it became free at the fore margin of 

 the induced sector (in the direction of the rotation), while at the 

 hind margin of the same, continually new positive electricity beiug 

 attached, relatively negative electricity became free. 



The positive electricity must, under these conditions, have over- 

 flowed from the fore to the hind margin of the sector, for which 

 there were in each ring two paths open, between which it must 

 have divided itself in the inverse ratio of their resistances. If the 



inducing sector comprises - of the circumference, the resistance 

 n 



of the path in the sector is to that of the path outside of it as 



l:ra— 1; and therefore of the current returns through the 



n to 



sector, and - outside of it. In the sector a quantity corresponding 



to the sum of the two currents is carried forward against the 

 current by convection. ' If, then, a convective motion of eletricity 

 acts like a conducted motion, the total motion in the sector is 



1 — -=-. But if the action of convective had been greater or 



n n D 



less than that of conducted motion, the excess, in one or the other 



direction, must have been shown on the sector. 



The experiments showed that, when the sector was small (g of 



