Striated Electrical Discharge. 529 



variable or intermittent, the density of ions will not have 

 sufficient time to take up its equilibrium distribution for each 

 small flow of current, and will therefore oscillate about a mean 

 which is intermediate between the theoretical distribution for 

 the maximum current and the distribution for zero-current. 

 Hence the graph for X or X 2 , if plotted from the readings of 

 an electrometer, will be intermediate between the graphs 

 found in this paper and a straight line. 



Thus intermittence in the current effects a smoothing out 

 of the graphs, and in particular, the minima of the graphs 

 will be much less marked. In connexion with this result, it 

 ought to be noticed that perfect steadiness of the current in 

 the part of the circuit outside the tube will not necessarily 

 involve the steadiness of the current at all points inside the 

 tube. 



§ 20. The results which have been reached have been 

 entirely qualitative. It would, however, be natural to inquire 

 within what limits of pressure, current, &c, a striated discharge 

 is possible, and what will be the length and number of the 

 striae corresponding to specified external conditions. To these 

 and all similar questions our theory has supplied no answer. 



If we refer back to the mathematical graph of fig. 5, we 

 see that given the pressure, the current, and the distance 

 between the electrodes, there are still an infinite number of 

 solutions mathematically possible, so that the original mathe- 

 matical solution can throw no light on the arrangement of 

 striae : all arrangements are equally possible. 



When w T e turn to the physical graph of fig. 6, this is no 

 longer the case. The three elements specified above will 

 •completely determine the graph ; there are no points in the 

 physical graph corresponding to the discontinuities in the 

 mathematical graph, at which a choice of ways is open to us. 

 This definiteness has been effected by the introduction of the 

 new intermolecular forces which are supposed to come into 

 play when the density of ions reaches a certain amount. Until, 

 therefore, we have a definite knowledge of the nature of these 

 forces, it will be impossible to arrive at an exact knowledge 

 of the relations between quantities which, except for these 

 forces, would be meaningless. 



In conclusion, I wish to express my deep indebtedness to 

 Prof. Thomson, not only for the continual use which I have 

 made of his published papers, but also for the assistance I 

 have received from him personally. 



