CORRELATION OF PHENOMENA 67 



out that this work scarcely touches the ground 

 which we propose to traverse, of its relation with 

 luminosity. Nay more, it seems as though the 

 tendency of the work of these few recent years has 

 been almost to ignore the views of the decade 

 that preceded them ; although some of these views 

 or hypotheses do not appear to have ever been 

 disproved, and it is to them that the facts and 

 speculations in this essay give a signal but perhaps 

 a qualified support. 



We do not refer to the ultra-atomic ; but what 

 may be called the mega-molecular theories of the 

 discharge of electricity through gases. 



The former theory, by which the corpuscle is 

 supposed, under the influence of the electric field, to 

 acquire sufficient energy, in moving through its free 

 path, to break up, or ionise, the molecules with which 

 it comes into collision, admirable as it is in afibrding 

 a simple explanation of many, if not most, of the 

 facts connected with the discharge of electricity in 

 gases, does not in itself give a clue to some residual 

 eifects. The small potential difierence in the positive 

 column, together with the intense luminosity in this 

 part of the discharge, favours, as it would appear, a 

 return to the older hypothesis of chains of molecules 

 in transmitting the current, and at the same time 

 giving rise to striae, and the brilliant glow ; not 

 the phosphorescent after-glow — although this also, 

 as we shall see, affords much evidence of great 

 molecular agglomerations at low pressure — but that 

 positive glow, which, though not persistent, is the 



F 2 



