TROWBRIDGE. — ELECTRICAL DISCHARGES. 451 



and a piece of tin-foil was then placed on the outer surface of this disk 

 B of paraffiiie and connected with a suitable insulated capacity. In this 

 case the fluorescent spot produced by electrostatic action was spread out 

 and ill defined, nauch as if a disk of glass of similar form should be 

 placed in the path of rays of light coming from a concave mirror. 



It therefore seems to us that in the geneiation of the X-rays the 

 electrostatic lines of induction are suddenly formed together with currents 

 of displacement, and that the wave front quickly advances, probably 

 with the velocity of light, and that the molecular effects observed are 

 merely the concomitants of the electromagnetic disturbances. 



Since the production of the anode rays requires a higher vacuum than 

 that at which the cathode rays are excited, we were interested to ascer- 

 tain if the molecules of the gas under their influence possessed a greater 

 or less velocity than that already found for particles moving from the 

 cathode. We accordingly constructed a tube which was substantially 

 similar to that employed by Prof. J. J. Thomson in his investigation 

 of the effect of the electrostatic field on the path of the molecules excited 

 by the cathode ray.* The modifications consisted in the substitution of a 

 continuous conductor A K C, Figure 14, for tlie cathode and anode of 

 his tube. In the figure, E and E' represent plates maintained at a known 

 potential, thus giving an electrostatic field, and Z) is a suitable diaphragm. 

 We expected to obtain a greater or less deviation of the position of the 

 fluorescent spot formed on the surface of the bulb by the anode rays 

 emanating from K. The inductive effect, however, of the plates E and 

 E' was so great when K was positive that no measures were possible. 

 This form of tube thus exhibited still more strongly the effects of elec- 

 trostatic induction, and confirmed us in the belief that electrostatic induc- 

 tion is an important phenomenon in the generation of the X-rays. 



Conclusions. 



1. Beyond one million volts the initial resistance of atmospheric air 

 to electrical discharges becomes less and less, and in certain conditions 

 can be as low as one thousand ohms between terminals two or three 

 inches apart. The curve therefore representing the relation between 

 length of spark and electromotive force departs from a straight line beyond 

 one million volts and approaches the axis which represents the increase 

 of voltage. 



2. When the initial resistance of highly rarefied air is broken down 

 by means of the X rays, it exhibits less resistance than it does at the 



* Phil. Mag., October, 1897. 



