22 J. E. Moore — Electrical Discharge from the point 



pheric pressure, to the total cessation of discharge at the 

 lowest pressures attainable. 



It is not intended in this paper to describe in detail the 

 constantly changing phenomena accompanying electrical dis- 

 charge, as the pressure of the gas is made continuously to vary. 

 It may be stated, however, that disruptive discharge ceases at a 

 pressure only a little below atmospheric, and, at a pressure of a 

 few centimeters of mercury, the discharged path is seen to be 

 broken up into a series of dark and bright striae. It is worthy 

 of special mention that the pressure at which this stratification 

 phenomenon makes itself apparent in any particular tube, 

 depends only on our ability to separate (optically) a bright and 

 a dark line, and that, with steady potential differences, one can 

 distinguish the striae much earlier, by using a proper magnify- 

 ing glass. This fact will be considered more fully farther on. 

 The distance between consecutive striae increases continuously, 

 while the bright and dark bands broaden until (in a tube with 

 fixed electrodes), the bright band of the last remaining stria 

 comes to coincide with one electrode, and the dark band with 

 the other. 



Almost immediately after the disappearance of the stratifica- 

 tion phenomenon, a column of blue light is seen to proceed 

 from the negative electrode. At its iirst appearance, this 

 cathode stream covers the whole surface of the negative elec- 

 trode, but, as the pressure of the gas is reduced, the area cov- 

 ered becomes less and less, until finally, at a pressure some- 

 thing less than one-millionth of an atmosphere, not only has 

 the stream come to cover a vanishingly small area of the nega- 

 tive electrode, but the blue light, by which the cathode stream 

 is characterized, has totally disappeared. 



With only an ordinary mercury pump we can reduce the 

 pressure of the gas much lower than that at which the cathode 

 stream disappears. On this further reduction of the gaseous 

 pressure, the fluorescence of the glass walls, as well as all other 

 evidence of electrical discharge, gradually vanishes. By absorb- 

 ing the residual gas by chemical or mechanical means, such low 

 pressures can be obtained that even though the electrodes are 

 only a few millimeters apart no discharge can be made to pass 

 between them, when their potential-difference is so great as to 

 be capable of causing a spark twenty-five or thirty centimeters 

 long in air. 



By reasoning on these limiting values of electrical discharge 

 precisely as we reason on the limiting values of functions 

 generally, it would seem to be abundantly evident that, in 

 space void of matter (that is in free ether), electrical discharge 

 through that space must be zero for all possible values of the 

 potential-difference between the electrodes. 



