450 PROCEEDINGS OP THE AMERICAN ACADEMY. 



terminal. By reversing the terminals it was then found that the breaking 

 down elfect was exerted principally upon the negative terminal. 



Since these investigations had shown that there are anode rays as 

 well as cathode rays, it was of interest to investigate the breaking down 

 effect of the anode rays. The form of tube was similar to those before 

 described, being provided with a continuous conductor, which could be 

 made positive or negative at will ; and it was found that the breaking 

 down effect was manifested also by the anode rays. 



With the tube represented in Figure 11, it was discovered that a differ- 

 ence of potential of 10,000 volts between GG' could produce the X-rays, 

 and indications of them could be obtained at 2,500 volts ; and it may be 

 that they exist even in discharges produced at a less voltage ; and is it not 

 possible that they may be the initial cause of the breaking down of even 

 atmospheric air, by setting up lines of polarization or stress ? 



The conviction has been expressed that electrostatic induction plays 

 an important part in the production of the X-rays, and the following 

 additional experiments were undertaken with a modified form of tube. 

 The first experimental tube was constructed as follows. A continuous con- 

 ductor A K C(Fig. 12) was enclosed in a Crookes tube. One portion of 

 this conductor consisted of an aluminium mirror, K. The wires leading 

 to K were covered with glass tubing, and the metallic surface of the back 

 of K was coated with an insulating glaze. The X-rays produced when 

 K was made the anode or cathode could be diverted in any direction 

 by means of a piece of tin-foil placed on the outside of the bulb and 

 connected with a suitable capacity, which was carefully insulated. A 

 metallic concave disk was then enclosed in the bulb. This disk, being 

 entirely disconnected from the continuous conductor A K C, could be 

 moved into any position by suitably inclining the tube. The disk B' 

 served to generate by induction a new beam of rays, which was much 

 intensified by placing a disk of tin-foil on the outside of the glass opposite 

 the disk inside the bulb and connecting this tin-foil with an insulated 

 piece of metal. 



The material of the enclosed disk did not appear to influence the 

 result. In one case it was made of lead, and in another of aluminium. 

 It was not necessary that the inner disk should be at the focus of K. A 

 number of such disks could doubtless be employed, all of which would 

 manifest such inductive action. 



In order to see if we could detect any refraction of the lines of elec- 

 trostatic force, a disk of parafhne having the same radius of curvature as 

 the outside of the glass bulb was placed at a certain position on the bulb, 



