Concentration Cells in Ionized Gases. 685 



Equilibrium*, and Mr. Jeans, in the paper cited in § 1, 

 consider second (and higher) order terms, and that they 

 are led to no inconsistency in supposing that co (i. e. K) is 

 stationary. I£ it is true that the case of surfaces of revolu- 

 tion is exceptional and that for them K^O,, then there will be 

 considerable difficulty in the examination of the stability of 

 these figures of equilibrium. 



In conclusion I desire to express my indebtedness to the 

 kindness of my colleague, Mr. G. H. Livens, for a careful 

 reading of the proofs of this paper. 



LXXI. On Concentration Cells in Ionized Gases. 

 By W. H. Jenkinson, B.Sc.(London Sf Sheffield)} . 



T^HE cause of the characteristic potential difference which 

 JL exists between a metal and air with which it is in 

 contact is not known with any certainty. It has been 

 supposed to be connected with incipient oxidation of the 

 metal, with the occlusion of gas in its surface, and also with 

 the corpuscular pressure in the interior of the metal. Apart 

 from any of these theories we may, however, draw conclusions 

 with regard to the potential step if we may assume that the 

 passage of electricity from metal to air is a phenomenon 

 which can be treated as reversible in a thermodynamic 

 scheme. Let us suppose that if a net transfer of Q units 

 of + electricity is made from a metal to air by sending an 

 infinitely small current for an infinitely long time through the 

 surface, then when it has been effected, .vQ units of -f elec- 

 tricity have passed from metal to air and (1 — x) Q units of 

 — electricity have passed from air to metal ; and that the 

 effect is reversible in the sense that it will be exactly annulled 

 if now Q units of — electricity are passed. We may also 

 make the further supposition that + and — electricity, both 

 in the air and in the metal, exist as entities exerting pressures 

 which are theoretically separately measurable. Then the 

 thermodynamical scheme which follows is exactly the same 

 as that originally worked out by Nernst in his theory of 

 electrolytic solution pressure for the P.D. between a metal 

 and an electrolyte, and it will follow in the same way in the 

 metal-air case that the potential step from metal up to air 

 must be given by the formula 



V = — (2#-l) log? i 



e n P 



* Phil. Trans. A. cc. (1902), pp. 251-314; and A. ccviii. (1908), 

 pp. 1-19. Collected Works, vol. iii. Paper 12. 

 f Communicated by Dr. S. R. Milner. 



