﻿Electrical Changes induced by Ultraviolet Ligld. 405 



of metal connected directly to the electroscope, as in fig. 2, 

 and with the plates in the position of a reflector at an angle 

 o£ 45°, as in fig. 1. The electroscope was charged negatively 

 in the former case, and positively in the latter ; on reversing 

 the charges, we were unable to get a measurable discharge. 

 It was found impossible to discharge the electroscope if any 

 of the metallic shields were used. This fact may easily account 

 for our being unable to discharge the electroscope rapidly 

 through an aluminium leaf, as Le Bon did ; for he used very 

 high tension sparks as his source of ultraviolet light, and 

 these are much richer in ultraviolet rays than the iron arc. 

 Elster and Greitel were, probably for the same reason, unable 

 to obtain any discharge with tin, cadmium, lead, copper, iron, 

 carbon, and platinum when illuminating with daylight. All the 

 measurements were made against a standard plate of copper 

 amalgam. It was found that plates of zinc amalgam and 

 tin amalgam gave discharges much too rapid for accurate 

 measurement; but copper amalgam was found to give constant 

 and reproducible values when prepared by placing a clean 

 piece of copper into a solution of mercurous nitrate for a few 

 moments, then removing, washing and drying; the plate 

 immediately before the measurement was polished with fine 

 emery-paper and finally cleaned with a dry piece of linen. 

 Such standard values were recovered for every substance 

 measured. For plates giving a very slow rate of discharge, 

 the electroscope was moved nearer the source of light, and a 

 new standard value taken. The constancy of the standard 

 plate was demonstrated by comparing it with a tin plate 

 under various conditions ; the measurements are given in 

 Table III. 



Table III. 



Positive charge 800 volts. 



Rate of discharge 

 of copper amalgam. 



Rate of discharge 

 of tin plate. 



Ratio Sn/Cu 

 amalgam. 



26 8 sees 



22-0 sees. 

 22-0 sees. 

 28 sees. 

 24*4 sees. 

 18'0 sees. 



0-82 

 0-84 

 0-835 

 0-841 

 0-82 

 Mean 0*83 



26-2 sees 



33*4 sees 



29-0 sees 



22-4 sees 





