10 



nature of the curves shows that they are produced by a mix- 

 ture of ions of varying mobility, the mean mobilities, how- 

 ever, being the same for positive and negative ions. 



No point of inflexion could be found in the mobility 

 curves, even when the field reached a value of 400 volts /cm. 



The shot used in these experiments and in the preceding 

 weighed '0392 gm. per shot. When glass beads (diameter 

 "036 in.) were substituted for the shot and allowed to fall 

 through a glass tube on a glass target the saturation curves 

 shown in fig. 8 were obtained. The character of these curves 

 shows that the variety of ions is still greater than in the 

 case of those produced by lead shot. Calculating for the 

 points P and Q as the elbows of the curves the mobilities are 

 found to be l'O x 10 ~ 3 for both positive and negative ions. 



The sign of the total charge on the air drawn over the 

 target was determined by drawing it through the Faraday 

 tube. When lead shot fell upon a target of any metal the 

 total charge on the air was small and of variable sign. On 

 the other hand, when the shot fell upon an insulator (glass, 

 ebonite, silk) the charge on the air was invariably positive. 

 The following are typical measurements : — 



Target. 

 Brass 





Eeading. 

 ± '2 



Zinc 

 Ebonite 





+ -2 ± -2 

 + 11 + -2 



Glass 

 Silk 





+ 6-9 + 5 

 + 10-0 (q.p.) 



When glass beads are used 

 negative : — 



Target. 

 Brass 

 Ebonite 



Glass 



the 



charge on the air is always 



Eeading. 



—10-0 + -1 



— 8-1 + -6 



— 4-5 



Silk 





— 30 



To test whether electrification incurred by the beads in 

 touching the metal tube in their fall had any influence a 

 glass tube was substituted for this, and was found to exert no 

 important influence on the readings. The ionization pro- 

 duced in the process of falling through the fall-tubes was also 

 measured, and found to be less than 10 per cent, of the total 

 ionization. 



In making experiments with the stream of lead shot fall- 

 ing upon a metal target it was observed that if unused shot 

 were allowed to fall upon a new target (say, of brass) the 



