Chemical Combination of Gases. 237 



each other in the way we have described will tend to alter 

 when the rings are placed in a region where the distribution 

 of Telocity is not uniform. Now the configuration may change 

 so that the rings tend to get closer together, or so that they 

 tend to separate: in the first case, as the rings get closer 

 together the effect of the disturbance will be to make the 

 connection between the rings firmer, but in the second case 

 the disturbance will tend to separate the rings. Now let us 

 suppose that we have two rings moving in the disturbed fluid: 

 if the effect of the disturbance be to make the one in front 

 move faster than the one in the rear, then the disturbance 

 will tend to separate the rings; but if the disturbance in the 

 fluid makes the one in front move more slowly than the one 

 in the rear, the effect of the disturbance will be to make the 

 rings cling more closely together. Now the properties of 

 vortex-motion are such that if the distribution of velocity in 

 the fluid is not uniform, then when two vortex-rings are 

 moving together in one direction the effect of the disturbance 

 in the fluid will be to make the ring in front move more 

 slowly than the one in the rear, while when they move in the 

 opposite direction the effects of the disturbance will be to 

 tend to make the ring in front move faster than the one in 

 the rear, and so tend to separate the rings. Thus we see 

 that it follows from our theory of chemical combination that 

 a molecule of a compound gas would tend to split up if it 

 moved in one way, while it would have no such tendency if 

 it moved in the opposite direction. If we take as the dis- 

 turbance that produced by an electric field, then it follows 

 that those molecules in the field which move in one direction 

 relatively to the lines of electric force will tend to split up 

 or be dissociated, while those moving in the opposite direction 

 will experience no such effect. This, taken in conjunction 

 with the theory of the electric discharge mentioned above, 

 will, I think, account for many of the differences between 

 the positive and the negative discharge ; in this paper, how- 

 ever, I shall confine myself to the more purely chemical 

 effects. 



We shall find it important to remember that there are two 

 kinds of dissociation: in the first the molecules are split up 

 by external aid, such as the action of light or electricity ; in 

 the second the body suffers dissociation at a high temperature 

 without there being any external disturbance. It is usual to 

 assume that the increase in the number of collisions is the 

 cause of the dissociation at high temperatures ; but when we 

 come to discuss the experiments on dissociation we shall, I 

 think, see reasons for believing that this is at any rate not 



