92 BELL SYSTEM TECHNICAL JOURNAL 



is further strengthened, the amplification becomes more intense, the 

 ionization more abundant, and there comes a point when the gas 

 "breaks down." A luminous discharge occurs, which may be transi- 

 tory (a spark) or durable (a glow or corona or arc). Breakdown and 

 the subsequent discharge occur even when there is no external agent of 

 ionization, apart from those feeble rays which constantly perv^ade the 

 atmosphere and every gas not shut off from the atmosphere by heavy 

 walls. Moreover, they occur with a high-frequency field, provided its 

 amplitude is raised to a sufficient value. This second part of the 

 present article is devoted to the conditions for breakdown by high- 

 frequency fields, and the characteristics of the discharge which sets in 

 thereafter. ^^ 



The discharge ensuing upon breakdown is as a rule enduring only 

 if the gas is rarefied (to a pressure not more than a few hundredths as 

 great as atmospheric) or one at least of the electrodes is sharply curved. 

 Otherwise, it is a spark. Striking as is the contrast between these 

 cases, one does well to disregard it while thinking about the processes 

 which may lead up to breakdown, or observing the conditions under 

 which this phenomenon occurs. What happens before the sudden 

 transition may be controlled by laws quite other than what happens 

 after it. Indeed, we know that the choice between spark and durable 

 glow-discharge is not so important in principle. The choice between 

 spark and glow is influenced, for instance, by the constants of the 

 circuit— not merely by the E.M.F. available, but also by the resistance 

 and inductance in series with the gas. It is advantageous, therefore, 

 to think of the conditions for breakdown and the presumptive details 

 of the process as forming a problem by themselves, apart from the 

 problems of the state which follows. 



General Remarks on Breakdown 

 Breakdown by "steady voltage" is brought about in either of two 

 ways: by gradually increasing the voltage across a pair of electrodes 

 separated by a stratum of gas, or by applying a fixed voltage and 

 gradually changing the distance between the electrodes. It is de- 

 tected either by the blaze of light attending the ensuing spark or 

 glow, or by a sudden violent change in the reading of a voltage- 

 measuring device shunted across the "gap," that is, connected across 

 the electrodes. The figure given as the "breakdown-potential" is the 

 last value of voltage recorded just before either of these events. 



i^' The order of treatment is thus the same as is customary in treatises on direct- 

 current phenomena, and as I have followed in my book "Electrical Phenomena in 

 Gases," to which again reference is occasionally made: first the drifting and accelera- 

 tions of electrons in gases exposed to weak fields, then the conditions for breakdown, 

 finally the laws of the luminous discharges ensuing after breakdown. Equations, 

 footnotes, and figures are numbered consecutively to those of Part I. 



