1891.] the Discharge of Lei/den Jars. 15 



Something apparently of the same sort has been quite recently 

 observed with sinuously alternating currents by Mr. Ferranti in the 

 Deptford mains. But whereas that case can be described as a long 

 stretch of capacity with locally concentrated inductance, mine is a long 

 stretch of inductance with, locally concentrated capacity. Accord- 

 ingly, while he observes an extra current-amplitude, I observe an 

 extra potential. 



The phenomenon in another form seems to have been first observed 

 by Sir W. R. Grove, and fully explained by Clerk Maxwell (see ' Phil. 

 Mag.,' for March and May, 1868). It was subsequently rediscovered 

 by Dr. Muirhead, and explained by Dr.-Hopkinson (' Journ. S.T.E./ 

 1884). A note sent by me to the ' Electrician ' for 24th April, 1891, 

 contains a summary of the history and explanation. 



DISCUSSION OP OVERFLOW AND SURGING EXPERIMENTS. 



17. For the complete explanation of the overflow experiments, the 

 static capacity of the long wire, and the momentum of the pulses 

 rushing along it, must be taken into account, and a wire is more 

 effective when insulated and charged than when lying on the 

 ground. 



It does act, however, even when lying on the ground, i.e., when its 

 magnetic momentum is all that can be supposed effective. But the 

 ordinary theory of discharge oscillation will not account for the jar 

 being thereby raised to a higher potential than it was at the begin- 

 ning of the series ; the amplitude of the vibration necessarily de- 

 creases. Hence it is probable that the fact of overflow does not 

 prove that the entire potential of the jar is raised; only that the 

 potential of the tinfoil edges is excessive. The charge is probably 

 not uniformly distributed at the extremity of each swing. The 

 fringe of sparklings above the edge of the tinfoil are x well known 

 whenever a jar is discharged ; and overflow is merely an exaggera- 

 tion of these sparklings, which usually leap up and subside. In fact 

 they can be seen to jump higher and higher, as the spark is gradually 

 increased, until the lip is leaped. 



The idea of the pulses rushing along the connecting wires, and 

 adding their momentum to the oscillation of the jar-discharge, sug- 

 gests that there must be a best length for the connectors, viz., when 

 the period of their pulses agrees with the period of oscillation of the 

 discharge ; and the fact that there is a best length is found experi- 

 mentally. 



The same length of connector is not equally effective with pint and 

 gallon jars. A longer one is best for the larger jar ; and if a con- 

 nector be too long it does not promote overflow any more vigorously 

 that if it were somewhat too short. 



