ELECTRICITY. 517 



cording to the force of the discharge. The most com- 

 mon case in which this effect is seen is in what is call- 

 ed the spontaneous discharge of a Leyden jar. When 

 a jar formed of thin glass, is very highly charged it 

 sometimes discharges itself through the glass itselt', pro- 

 ducing the appearances above described, with this addi- 

 tion, however, that the tinfoil is protruded, both on the 

 inside and outside, forming in both cases a large and 

 conspicuous bur. This result seems to conlirm the 

 opinion stated above, that these protrusions are owing 

 to the explosive force of highly rarified air, or to some 

 similar cause, rather than to the impetus of the fluid, for 

 the glass, at the plar.e of fracture, is completely pulver- 

 ized, but it is not displaced, yet the tinfoil is displaced, 

 there being between it and the glass a quantity of air 

 mingled with the paste or other cementing substance. 



' The expansion of air by the passage of the electri- 

 cal fluid, either in the form of sparks or shocks, is shown 

 in the following experiment of Kinnersley, the appara- 

 tus for which has been called the electrical air thermom- 

 eter. It consists, of a glass tube closed at both ends by 

 air-tight brass caps, through which two wires slide in 

 the direction of the axis of the tube. These wires are 

 terminated by brass balls, which are made to approach 

 within the striking distance. To an aperture in the bot- 

 tom of the lower cap, is fitted a bent tube of glass which 

 turns upwards, and is open at both ends ; the bent part 

 is tilled with mercury, or with a colored fluid, which may 

 indicate by its rising or falling in the tube, any dilata- 

 tion or contraction that may take place in the air 

 within the vessel. It is found that every time a spark 

 passes between the brass balls, the fluid suddenly rises, 

 but descends again to its former level immediately after 

 each explosion ; thus showing that the dilatation of the 

 air, produced by the abrupt passage of electricity, is but 

 of momentary duration. 



' When a strong electrical charge is sent through a 

 very confined portion of air, the explosive effects produc- 

 ed by it, are as considerable as those we have seen ex- 

 hibited by denser fluids. Thus if a piece of plate glass, 

 of the size of a square inch, and half an inch in thjck- 



