418 Mr. G. W. Walker. On the Drift produced in [Dec. 16, 



Case II. Let x = -\uil and z .= initially. 

 Thence a = - %ud and c = V. 



In this case 



x = exactly. 

 Thus the apparent velocities are 



x = exactly, 



a>2d2 d 



~ 



Thus the ion will seem to move in the direction from which the 

 impulses come. It is worth while to note that the x velocity 

 vanishes, and so the ion will drift backwards without altering its x 

 co-ordinate. 



In both these cases the initial circumstances are such that the iori 

 succeeds in getting through the first pulse. It will be seen that the 

 initial circumstances can be so chosen that it fails to do so. This, how- 

 ever, involves the result that at some point of the circular path I = 0, 

 or in other words that the ion is moving with the velocity of the 

 waves. Now the equations break down before this point; but the 

 result may be held to indicate that if the ion is originally moving in 

 the direction z with a velocity a little less than V, it may, so to speak, 

 be picked up by the waves and carried forward with the velocity V. 



These cases are sufficient to illustrate the general feature, and it may 

 be noted that the apparent x velocity is an odd function of the charge 0, 

 while the z velocity is an even function of the charge. This last result 

 leads us to expect that even a neutral molecule made up of positive 

 and negative ions will also be made to drift in the direction in which 

 the waves are travelling. 



We thus arrive at the conclusion that the propagation of plane 

 polarised disturbances through a portion of space containing ions 

 involves drifting of both positive and negative ions which may be with 

 or against the direction of propagation according to the initial circum- 

 stances. Since the z motion does not depend on the orientation of the 

 plane of polarisation, similar results must follow for unpolarised 

 disturbances. 



The restoration of the initial velocities relative to the fixed origin, 

 after the passage of what we may call a complete pulse, shows that no 

 energy (relative to the fixed origin) is permanently abstracted by the 

 ions, although during one portion of the pulse energy is abstracted 



