Discharge in Rarefied Gases* 181 



The dark space does not represents has been often assumed, 

 the prolongation of the negative rays which have lost in 

 brightness in consequence of their expansion. The negative 

 rajs have the property of rectilinear radiation and are termi- 

 nated by a solid wall] they cannot, therefore, turn a corner. 

 The appearances obtained with the discharge in bent tubes, 

 such as figs. 5 and 6, require therefore no further explana- 

 tion to show that the dark space cannot be considered as a 

 continuation of the kathode-light, and that it possesses itself 

 no property of rectilinear radiation. 



If, therefore, we assume that the current of the kathode- 

 light propagates itself to the first positive layer, it is necessary 

 to suppose that the current between the two is changed for a 

 certain distance into a new form. 



Returning to the straight tube (fig. 4), if we continue the 

 exhaustion further from the point at which the dark space 

 makes its appearance, the positive layers retreat slowly towards 

 the anode ; at the same time the rays of the kathode-light 

 lengthen, and more rapidly than the positive layers retreat. 

 We thus reach a point at which the dark space disappears, 

 and the negative light reaches to the first layer of the positive 



Kght. 



Now we should be obliged to assume that the new form of 

 conduction has disappeared, although in the visible portion of 

 the discharge (neglecting the small displacements of the posi- 

 tive layers) no change has occurred in the meantime except 

 that the negative layers have lengthened; their properties, 

 like those of the positive layers, are altogether the same as 

 before. 



If now we exhaust still further, the positive layers retreat 

 still further; the rays of the kathode-light lengthen still more, 

 and again more rapidly than the positive layers retreat. The 

 negative light now penetrates into the layers, whilst its proper- 

 ties remain unaltered, loithout confusion with those of the 

 positive light into which it penetrates. 



The proof of the penetration of the negative light into the 

 positive light may be obtained in various ways. In fig. 7, 

 which represents the section of a vessel consisting of three 

 cylinders, the kathode a is the section of a thick wire covered 

 on the sides with glass melted round it. If the wire c near a 

 is made the anode, there is no manifestion of positive light 

 except in the immediate neighbourhood of the anode, and only 



light and the wall of the tube is a dark space, in wide tubes of several 

 millimetres breadth, not mentioned in previous descriptions. In order not 

 to make the figure of inconvenient length, the third layer of the kathode- 

 light in fig. 4 is shown of less thickness than is actually the case. 



