198 TROWBRIDGE. — HIGH ELECTROMOTIVE FOKCE. 
proportional to the voltage; beyond this point leakage and electrostatic induction 
prevent accurate measurements. In order to study high voltage the apparatus 
should be lifted outside a building and placed high above the ground. 
It seemed evident from observation of the phenomena that air at atmospheric 
pressure breaks down with great facility under high voltage combined with large 
amperage. The resistance of the electric spark varies greatly with the amount of 
current carried by the spark ; and the non-concordant results obtained by different 
observers for this resistance can be explained by this fact. In one case I found that 
under a difference of potential of three million volts the spark passed through five 
centimeters of air in preference to a circuit through a solution of sulphate of copper 
of one thousand ohms resistance. 
One is impressed in studying high voltage combined with large amperage that the 
study of electrical discharges by means of Holtz machines or other forms of glass 
inductors leads to limited conceptions of the amount of energy in lightning discharges. 
If Benjamin Franklin had worked with a high-tension storage battery, he probably 
never would have dared to try his celebrated kite experiment. Experience has shown 
that even five hundred volts combined with large current is sufficient to cause death. 
The apparent diameter of the electric spark and its apparent direction are deter- 
mined largely by physiological considerations. 
To the eye, sparks of six or seven feet in length appear of at least a tenth of an 
inch in diameter. To obtain an idea of the size of the thread of luminosity, the 
discharge was passed through the finest needle hole in a plate of glass five feet square. 
This hole was made by plugging a much larger hole bored in 
and piercing the paraffine with a needle. When the spark 
.- 
paraffii 
opposi 
the hole, each one a foot and a half from it, the spark sought the hole and passed 
through it. A photograph shows an apparent breadth of spark much greater than the 
diameter of the hole. Possibly only a portion of the spark passed through the hole, 
for a surface ionization may have taken place over the surface of the glass. 
In regard to the apparent direction of a lightning discharge I tried the foil 
o 
experiment. One observer changed the poles of the apparatus for discharging the 
condensers, while another, looking through an opening which concealed the spark 
and only revealed the middle portion of the discharge, noted d 
impressions of direction. On comparing the notes of the two experimenters, i 
ment was found in regard to direction. This was also true when the disch 
made non-oscillatory. 
rati- 
o 
Considerations of direction may arise from difference of luminosity and difference 
