of a Magneto or Induction- Coil. 487 



is soon reached in which the potential is nearly constant for 

 a considerable time, longer with the slower separation ; and 

 that subsequently the potential rises rapidly and attains a 

 maximum, nearly independent of the speed of separation, 

 at a time after the beginning of the rise which is again 

 nearly independent of the speed of separation and also nearly 

 equal to the time that curve A takes to rise from zero to 

 its maximum. 



The form of the curves is exactly what we should expect 

 if an arc is formed, and utterly unlike that which is 

 characteristic of the passage of a spark (see p. 492). The 

 initial rate of rise of potential is small because the current 

 is not broken when the contacts separate but is maintained 

 by the arc between the hot contacts. The constant potential 

 which persists for some time is very nearly that of the 

 battery supplying the primary current (10 volts). At the 

 point where the second rise of potential begins the arc ceases, 

 and the rest of the curve represents approximately that 

 which would have been obtained if a current, equal to that 

 flowing in the arc, had been broken at the first separation 

 of the contacts. 



Many more experiments would be necessary to establish 

 the complete accuracy of ail the statements that have been 

 made and to work out in detail the quantitative relations 

 which are suggested. But the observations which have 

 been described are sufficient to show that the discharge 

 which occurs when the " flash " is seen at the broken 

 contact is of the nature of an arc rather than a spark, and 

 that it is conditioned by the heating of the electrodes. 



And it is now apparent why the "flash " may be prevented 

 by decreasing the rate of rise of potential, but not by a 

 corresponding increase in the rate of separation of the 

 terminals. The potential of the battery (or, in the case of 

 the magneto, that due to the rotation of the armature in the 

 magnetic field) is not sufficient to maintain the arc per- 

 manently. If this potential is only 10 volts, then, however 

 slowly the contact is opened, the current ceases a very 

 small fraction of a second after the distance between the 

 contacts has reached a value which is certainly less than 

 0*01 mm. The energy dissipated in the discharge is not 

 sufficient to maintain the electrodes at the high temperature 

 necessary for the arc- discharge. On the other hand, the 

 passage of the discharge and the dissipation of energy which 

 accompanies it delay the cooling of the electrodes, and they 

 delay it the longer the greater is the potential difference 

 maintaining the discharge. Accordingly the more slowly 



