sOMl- rO.V7/;.UrOA'./A'r .(/)C./.V( /V /.V /7/)S7( V 17 II.! 



Ill .1 l>,ittfry to the ends of a wiro, and promptly llu- i-k-ctric potential 

 (li>triliutes itsflf with a uniform gradient alon^ the win- and a ciirriMit 

 flows steadiK' down il. So rijiorously is llu- ciirri'iil jiroporlional lo 

 till' \(illai;e lielween the ends of the wire, over \ er%- wide ranges of 

 \olta^e .md eurrenl, thai we rejjard the ratio as an essential constant 

 of the wire: and we regard the ratio of potential-gradient (elertrh- 

 held) lo current density as an esseiilial characteristic of the metal, 

 and gi\e it a name — resistivity or specific resistance -and refer to 

 theories of conduction in metals as theories of metallic resistance. 

 1 1 all seenis exceedingly simple, and \et in the forejjoing article of this 

 series I have shown how all the atlempls lo interpret it have ji;one 

 in vain. Much more complex in appearance is the discharjje throuj^h 

 a gas. One applies the terminals of a battery to a pair of elect rotles 

 facing one another in the opwn air, and perhaps nothing happens, or 

 so minute a current flows that the most delicate of instruments is de- 

 manded to detect it; and then when the I)attery-\oItagc is very slightly 

 raised, there may be an explosion with a blaze of light, dissociating the 

 gas and corroding the electrodes, and draining off the a\'aiiablc elec- 

 tricity in a moment. Or if one of the electrodes is acutely pointed 

 there may be glows and luminous sheaths around it or tentacles of 

 bluish light ramif\ing from it far and wide through the air. Or the 

 ilischarge may rise to the heat of incandescence, Tind the gas and the 

 electrodes shine with a blinding radiance, the brightest light that can 

 tie kindletl on the earth. Or if the electrodes are enclosed in a tube 

 containing a rarefied gas or vapor, the gas flares up into an extraordi- 

 nar\' pattern of light and shade, lucent vividly-colored clouds floating 

 Itetween regions glowing feebly or obscure; and as the gas is gradually 

 pumped awa>', the pattern changes and fades, a straight beam of 

 electrons manifests itself by a luminous column traversing the tube, 

 the glass walls flash out in a green fluorescence, and finalK' all becomes 

 extinct. As for that e\en gradient, and that constant proportion 

 between current and field strength distinguishing the metals, we 

 cannot find them here. There is no such thing as the resistance of 

 a gas; we had better forget the word, we cannot attach any physical 

 meaning to the ratio of current and voltage. 



I must not give the impression that all these manifold forms of the 

 electric discharge in gases are understood. Certain of the simplest 

 of them have been clarified, and as a result still simpler ones have 

 been realized and comprehended in their turns, and so on down to 

 the simplest of all, which is the discharge across a vacuum. This 

 sounds somewhat like a paradox and so it would have seemed thirty 

 or forty years ago, when electricity was thought to be inseparable 



