Mass of the Ions in the Electric Wind in Air. 417 



Mass of the Ions. 



Equation (v.) contains the symbol it, which stands for that 

 part of the pressure resulting from the momentum of the 

 ions themselves. If F represents the strong field close to the 

 point in which the ions start, e the ratio of the mass of the 

 ion to the charge it carries, and C the current from the point, 



Jbr=C«VF, 



where k is 1 if the ions all start in the direction in which the 

 pressure is measured, 2 if they radiate uniformly along the 

 normals to the hemispherical surface of the point, and greater 

 than 2 if any of the discharge takes place from the sides of 

 the point. For a weak current, or for a point in a strong 

 parallel field, k will thus lie between 1 and 2. 



The positive curve in curves III. cuts the line of dynes 

 above the origin. Calculating the position of the cutting- 

 point from the seven lowest readings gives 7r=0 , 76 dyne ; 

 or, by taking all the readings except the highest, 7r = 0"G6 

 dyne, k is unknown; say 1*5. C = 7100 e.s. units. V=413. 

 F for a point of diameter 0*0055 centim. =1870 e.s. units*, 

 hence 



e = 7100x413°x 7 1870 = X ' 9 X 10 ~ 10 *'»' Units ' 



For oxygen the electrolytic value of e is 2*8 x 10 -H e.s. 

 units. On Rutherford's hypothesis of molecular clusters this 

 would mean that the diameter of each cluster contains 

 Vl-9xl0- 10 ^-2-8xl0- 14 = 19 ordinary molecules in a row. 

 Rutherford himself estimates the number at 5'5 for air ; 

 his calculation being based on considerations of viscosity. His 

 result, however, really means that the diameter of the sphere 

 of action of a molecular cluster is 5*5 times that of a single 

 molecule of air, which is not necessarily the same thing. For 

 it seems unlikely that the molecules forming a cluster do not 

 come nearer than just to feel each other's influence ; and if 

 they do, there must be more than 5"5 in the diameter. The 

 two results are therefore not inconsistent. 



With the ring and point apparatus so far employed ir was 

 not obtainable. I therefore took a larger and thinner ring of 



* This is calculated from a formula which I found for the field at a 

 point just on the verge of discharging, as I have since satisfied myself 

 that this field remains nearly constant while the discharge is occurring. 

 The proof of this I hope to publish shortly in connexion with other work. 

 The formula is F = 10oX r-' 08 , where /• is the radius of the point in centi- 

 metres, F the field close to the point, and air the gas ; the discharge being 

 positive. For negative discharge divide by 1-3. (Phil. Mag. Sept. 1891, 

 p. 293.) 



Phil. Mag. S. 5. Vol. 48. No. 294. Nov. 1899. 2 G 



