176 Rev. P. J. Kirkby on Union of Hydrogen and Oxygen 



Before proceeding to discuss the tables in detail, I may 

 point out one remarkable result which they establish, viz., 

 the substantial constancy of Ap/AQ 



Now it is easy to show that 4 - % is very nearly the number 



AQ 



of molecules of water- vapour which result from the formation 



and passage to the electrodes of each pair of ions. For if N 



is the number of molecules in a cubic centimetre of a gas at 



15° C. and 760 mm., then in an apparatus of S c.c. capacity, 



a fall of pressure Ap results from the disappearance of 



N x S x A/> 



„,..-, molecules. 



( oU 



Again, since one coulomb contains 3 X 10 9 electrostatic 

 units, AQ coulombs is carried to an electrode by 



AQx 



ions, if e is the electrostatic charge on each. 



Therefore, as each pair of ions passes to the electrodes, 



^-77{ — r> — v^M \w\ molecules of the gases disappear. 

 7b0x3xl(r J x AQ ° 



and 



~Ne=l'22 x 10 10 (Townsend, Phil. Trans. 1899), 

 S =1130 c.c, nearly. 



A/> 

 Therefore the fraction last written is almost precisely 6^. 



Thus the formation of l.f molecules of water-vapour is 



associated with the passage to the electrodes of each pair 

 of ions. 



Now the tables show that, in spite of great variations in 

 the pressure and in the distance between the electrodes, the 



value of —y. is nevertheless always about 2. It varies between 



1'6 and 2*4; but it is more nearly constant than these limits 

 indicate. 



It follows then, that with parallel plane electrodes, whatever 

 their distance apart (at least, so long as it falls within the 

 range defined by the tables), and for all such small pressures 

 as appear in the tables, from 6 to 9 times as many molecules 

 of water-vapour are formed as pairs of ions. This result is 

 independent of the capacity of the apparatus, and, as will be 

 seen presently, of the nature of the electrodes. 



