Observations on some Experiments in Electricity. 67 



at the greatest possible distance, — that is at the whole length of the 

 diameter, — from each other, the intensity is at a maximum. To 

 prove this we have only to employ the common plate machine with 

 four rubhers on which the collectors are placed at two opposite points 

 on the circumference, each a quadrant from one pair of the rubbers. 

 A machine thus arranged yields a very copious supply of electricity 

 and will charge a battery with great rapidity, but owing to the proxi- 

 mity of the rubber to the collector, the charge is in some measure 

 condensed,— the stratum of air between the two, serving a purpose 

 analogous to that of the glass in the Leyden jar. If however we re- 

 move the collecting points from their usual position, and also detach 

 one pair of the rubbers and substitute a collector in their place, we 

 shall discover, at once, a remarkable difference in the working of the 

 machine, while it furnishes a far less rapid accumulation of electricity 

 in a battery, the sparks will be greatly increased in intensity. By a 

 change such as is here described, I have caused a machine, which 

 was yielding sparks of only three inches, to extend them immediate- 

 ly to eight inches and eight-tenths in length, the conductors, balls, and 

 insulators remaining, in both cases, the same. 



9. A method of producing rotation. 



The following experiinent furnishes an illustration of continued ro- 

 tary motion -derived from electricity, different from that q( Jiies with 

 recurved points. 



Near the lower edge of a vertical plate machine four feet in di- 

 ameter, of which the rubbers are on a level with the axis, I place on 

 an insulating stand, a sharp pointed pivot, to receive a brass wire, or 

 needle, five inches long which may revolve freely in a horizontal 

 plane, like a common compass needle. The wire is furnished with 

 a hollow brass ball at each end, half an inch in diameter. This nee- 

 dle, of course revolves in a plane at right angles to that of the plate, 

 and is placed so far only from the latter that the balls will not inter- 

 fere with it when the two revolve simultaneously. For the purpose 

 of this experiment the collecting points and prime conductor. of the. 

 machine, ordinarily placed at the vertical points of its periphery, are 

 removed. On turning the plate the needle with its balls begins at 

 once either to revolve, or to oscillate through considerable arcs. In 

 the latter case the oscillation is in a short time converted into a com- 

 plete rotation, that soon increases in rapidity to two or three times a 

 second, according to the action of the machinCo Now if a fly-wheel 



