CONTEMPORARY ADVANCES IN PHYSICS 111 



attainment of another and much lower constant value. There the 

 critical frequency was found to depend on the gap-width and on the 

 curvature of the electrodes; here, the data are too scanty to permit 

 of a similar, or any conclusion. The behavior of the breakdown- 

 potential Vsm in these tubes of rarefied air is of the same sort but 

 (according to Rohde) the percentage-drop from its low-frequency 

 value to its value at the highest frequency attained is much less 

 striking than that of the minimum maintaining potential Vmm- The 

 ratio of Vmm to Vsm therefore decreases with increase of v, descending 

 for argon to the value 0.1, for mercury to the fantastically low value 

 0.036. 



The smallness of these lowest values of the maintaining potential is 

 something extraordinary. They are, of course, much smaller than the 

 minimum maintaining potential of the direct-current glow, which is 

 the cathode-fall, and is of the order of hundreds of volts. Now, the 

 office of the cathode-fall is to maintain the outflow of electrons from 

 the cathode (this is proved by the fact, among others, that it becomes 

 dispensable if the cathode is heated to such a degree that the outflow 

 becomes spontaneous). The conclusion therefore is, that in the high- 

 frequency discharge the demand for electrons from the electrodes is 

 minimized if not abolished. Even so, the minuteness of the voltage- 

 amplitudes remains astonishing. Taking Rohde's data for the fre- 

 quency of 10^ and going from the least toward the most striking case, 

 we notice: air 14 volts, oxygen 12, nitrogen 12.5, hydrogen 15.5, 

 helium 16, neon 11, argon 8, mercury 5 volts, I illustrate this by 

 Rohde's curve (Fig. 23) of maintaining-potential versus pressure for 

 neon, though in one respect the curve is quite untypical: no other gas 

 exhibits so long a nearly horizontal arc (in a tube of 24 mm. diameter, 

 and 19 mm. from one to the other of the electrodes). 



More striking yet are some of the values obtained by C. and H. 



Gutton, whose flock of curves of F^-vs-^ for various frequencies and 



Fm-vs-f for various pressures, obtained in long tubes with rarefied 



hydrogen within and metal electrodes outside, is almost as intricate 



and perplexing as the family of curves for the breakdown-potential 



of which I spoke above. Many indeed exhibit no minimum at all. 



However, with a tube 5.3 cm. long he maintained the discharge, 



at some 4-10^ cycles, with a voltage of amplitude 5.7; and with a 



twenty-centimeter tube at 2- 10^ cycles he kept it alive with a voltage 



amplitude of 40, which considering the length is almost equally 



remarkable.'^" 



^'' In consulting papers of the Guttons, remember that they give R.M.S. values 

 of voltage and fieldstrength, not peak-values nor amplitudes. 



