1896-97.] Influence of Ront gen Bays on Electric Conduction. 403 
B had retained its charge practically undiminished, and that A had 
acquired a very slight charge of the opposite kind. 
§11. Instead of placing the two ends of the paraffin in imme- 
diate contact, four pieces of metal of T \y of a mm. thickness were 
placed one at each corner of one of the ends, so that when the 
sticks of paraffin were placed end to end there was now an air 
space of t 3 q of a mm. between the paraffin ends. When B was 
charged and A not charged, and the two put end to end, and 
then exposed to flame or to Rontgen rays, it was found that B’s 
end still retained its charge, and A’s end acquired a very slight 
opposite charge. 
With an air space of \ of a mm. the same results were 
obtained. 
With the air space increased to 1 mm. the charge on B was less 
after the two had been passed through the flame or the rays. 
§ 12. Similar experiments were made with rods of glass and 
of ebonite, with similar results. 
Art. III. — On the Influence of Rontgen Rays in respect 
to Electric Conduction through Air, Paraffin, and 
Glass. 
(Read March 1, 1897.) 
§ 13. We have in §§ 5 to 10 described experiments respecting 
electric conduction through air, paraffin, or glass, when Rontgen rays 
fall on metal surrounded by air, paraffin, or glass, and positively or 
negatively electrified to potentials of two or three volts. We found 
that although air is rendered conductive, paraffin and glass are not 
rendered sensibly conductive when the differences of potential con* 
eerned are not more than two or three volts per centimetre of air, 
or per centimetre of paraffin, or per half-millimetre of glass. 
We have now to describe an extension of the investigation to 
much higher voltages, in which we use an arrangement of two 
(quasi) Leyden jars, A and B, with their inside coatings con- 
nected together. The outside coating of A was connected to 
sheaths, the outside of B to the insulated terminal of the electro- 
