594 Mr. G. C. Simpson on 



which the rings came, no charge could be observed on the 

 rings entering the cage ; but on reducing the distance 

 between the orifice and cage a distinct positive charge was 

 obtained from the rings. Thus during the time (about a 

 tenth of a second) the ionized rings had been passing through 

 the unionized air on their way to the cage, a diffusion of 

 the ions had taken place ; but as the rings entered the cage 

 with a positive charge we may safely say that the negative 

 ions had diffused away the more rapidly. The fact that 

 there was no charging of the cage when the rings had to 

 travel further thnn 20 cms., showed that within one-fifth of a 

 second both positive and negative ions had entirely diffused 

 away. In this way it was possible to show directly that 

 negative ions have a greater coefficient of diffusion than 

 positive ones. 



Two interesting observations showing charging as a result 

 of ion absorption must be here described. The first was 

 observed when working with the ionized vortex rings. It 

 was then found that when the box from which the rings Avere 

 sent out had been filled with tobacco-smoke in order to make 

 the rings visible they entered the cage with a negative 

 instead of a positive charge. 



The second observation was as follows : — Fine white sand 

 w^as allowed to run out of a metal funnel into a saucer some 

 distance below. As a result of the friction between the metal 

 and the sand, the latter was positively charged. A 

 ring wrapped round with cotton-wool soaked in spirits was 

 now so placed that the streau] of sand passed through it on 

 its way from the funnel to the saucer. On lighting the spirits 

 the sand passed through a ring of flame. It was then found 

 that the sand carried a neo-ative charo-e to the saucer. At 

 first it was thought that this change in sign was due to a 

 change in the frictional electricity, owing to the funnel 

 becoming hot, but that this was not so could be seen by 

 removing the flame, when at once the positive charge re- 

 turned although the funnel had had no time to cool. 



The explanation of these two observaiions is the same. 

 The smoke ring on leaving the box consists of a mass of carbon 

 particles mixed with negative and positive ions. The sand 

 on falling through the flam.e carries with it a quantity of 

 the highly ionized gas of the flame. Thus in both cases 

 we have to do with a mixture of material particles with 

 positive and negative ions. As soon as this mixture enters 

 ordinary air diflusion takes place, but the ions in the middle 

 of the mass have to diffuse through the material particles. 

 We may consider the process as being similar to a diffusion 



