608 Prof. J. 8. Townsend on the GencsU of Ions 



Table VII. — Hydrogen, pressure 1 mm. 



r/ 



2. 



4. 



6. 



7. 



a 



10. 



1 





X = 262 q 



« = 2-75 

 ;3=:-202 ^^ 



1-7<J 



3-2 



(5-4 

 6-4 





16-3 

 lft-2 



180? 

 196 



1-79 



3-2 





X=3oO q 



«=31 



/3 = -33 '' 



1-92 



4-0 



10-4 



24-5 







1-92 



40 



10-6 



23-8 



The hydrooen which was used in these experiments was 

 very carefully prepared. The apparatus was tested before 

 the hydrogen was admitted and it was found that when 

 exhausted to about a twentieth of a millimetre, the increase 

 of pressure during a day was only one three-hundredth of a 

 millimetre, so that any impurity arising from leakage or air 

 coming off the sides of the apparatus must have been very 

 small. The hydrogen was generated by electrolysis from a 

 solution of caustic potash and a special form of apparatus was 

 used so that no small bubbles of oxygen should get mixed 

 with the hydrogen. While the gases were being evolved the 

 solution was kept hot in order that no gas should be absorbed 

 by the solution and be thus conveyed from the neighbourhood 

 of one electrode to the other. The hydrogen was kept at 

 reduced pressure for 24 hours in an airtight receiver con- 

 taining phosphorus pentoxide before it was admitted to the 

 vessel containing the parallel plates. The apparatus was filled 

 to a pressure of 10 centimetres with dry hydrogen and 

 exhausted to a pressure of one millimetre several times so as 

 to eliminate traces of air. In order to reduce as far as 

 possible any error arising from the small leakage or air 

 coming off the sides of the glass a fresh supply of hydrogen 

 was introduced a short time before a set of experiments were 

 made. 



It will be seen that the values of n derived from the formula 



n = -7^4 — w\ — aoree with the numbers found for a experi- 



mentally. Since g is expressed in arbitrary units it is possible 

 to find values for a and /3 which give values of n equal to 

 values of g at three different distances. It is evident that the 

 formula gives the correct conductivities for a large range of 

 distances. 



6. Further considerations in fav-oau* of the theorv can be 



