TROWBRIDGE. — ELECTRIC DISCHARGES. 631 



pole ; the intervals between the masses acting like cathode spaces, the 

 number of luminous masses increases with the strength of the current 

 passing through the tube. They seem to be shot slowly out of the posi- 

 tive pole, and in their movement constitute a very beautiful experiment. 

 The accompanying table represents this increase of the masses with 

 increase of current. 



It would seem that the chain of positive and negative intervals with 

 very powerful currents becomes a chain with infinitely small intervals 

 between the positive and negative masses. Perhaps this is also the 

 mechanism of the electric discharge through solid conductors. The 

 pressure in the tube varied from four centimeters to half a centimeter. 

 With lower pressures the ordinary form of striae was obtained. With 

 narrow tubes the curvature of the striae increased, and the striae were 

 apparently firmly held by the walls of the tube, recalling the phenome- 

 non of the liquid meniscus. If we suppose, according to Townshend's 

 hypothesis, that the striae indicate the collision of positive and negative 

 ions, this collision would probably be modified by changes of velocity of 

 collision near the walls of the tube in much the same way that the 

 molecular actions at the surface of a fluid are modified by the attracting 

 forces of the containing tube. 



I believe that the slow movement observed by Rhigi is an ionization 

 change produced by a change of pressure resulting from the heat of dis- 

 charge of a condenser, combined with a change in electric potential. It 

 has nothing to do with the velocity of the electric impulses, and cannot 

 be called a slow movement of electricity. It seems probable that globe 

 or ball lightning, and other apparently slow movements of luminous 

 effects observed in thunder-storms, are of the nature of the ionization 

 effects described in this paper. 



The ionization theory of Townshend to my mind is the best explana- 

 tion of these slow moving luminous effects. One can conceive of a non- 



