'.'I Prof. J. , I. Thomson. On th? Propagation of the [Jan. 15, 



the commencement of the dissociation of a molecule to the end of the 

 recombination of its atom with those of neighbouring molecules, a 

 positive atom will hnve disappeared from one end of the chain and 

 appeared at the other. Thus in this case, since the time taken for the 

 decomposition and recombination of the molecules is independent of 

 the length of the chain, whatever the length of the chain may be, th; 

 positive charge will travel from one end of the chain to the other in the 

 same time, and thus the velocity of the discharge will be proportional 

 to the length of the chain. In the paper referred to above, it i> 

 gested that the high velocity of the discharge ( f the positive column 

 is attained by the formation of Grotthns chains of suitable length, 

 the column thus consisting of a series of separate discharges, the 

 length of each discharge being that of the Grotthus chain ; these 

 separate discharges are made manifest in the stratification which is so 

 striking a feature of the positive column, the space between the 

 bright portions of two striae corresponding to the length of the 

 Grotthus chain ; thus, on this view, the stratifications are the mani- 

 festations of the machinery which enable the positive discharge to 

 travel at snch a rate. In the paper in the ' Philosophical Magazine ' 

 it is shown that this view of the discharge agrees well with what is 

 known as to the behaviour of striae. 



The preceding experiments show that the tubes of force which we 

 imagine as stretching round the circuit, and contracting when the 

 discharge takes place, are anchored almost completely to the negative 

 electrode. When the discharge begins to pass., the ends of these 

 tubes near the positive electrode will be agitated in an approxi- 

 mately periodic way, electrical vibrations will travel along the tubes 

 with the velocity of light, and, as one end of the tube is fixed, these 

 will form stationary vibrations ; these stationary vibrations may be 

 conceived to give the molecules of the gas in the tube a certain 

 periodicity of arrangement, and lead to the formation of the Grotthus 

 chains of definite lengths, required by the preceding explanation. It 

 will be seen that this would make the position of the striae depend on 

 that of the negative electrode, so that when the latter is moved the- 

 strire ought to be displaced ; this effect has been observed by Gold- 

 stein. 



As an alternative to the preceding view, it might perhaps be 

 urged that the luminosity of the positive column outruns the positive 

 discharge. This view, however, seems to be quite untenable in the 

 face of Spottiswoode's and Monlton's experiments on the sensitive 

 st.ite of the electric discharge (' Phil. Trans.'), for the relief effects 

 observed in their experiments seem to show, without ambiguity, 

 that the positive luminosity is coincident with free positive elec- 

 tricity. 



Wo have also no evidence that a gas can be made luminous by 



