22 Dr. E. N. da C. Andrade on the Electrical 



from the deflexion of the streak (of the given brightness) is 



A A . cm. /volt 



0"0;> — / . 



sec./ cm. 



The rough agreement of: these numbers is a satisfactory 



confirmation. 



§ 5. Relation between Conductivity ant/ Light-emission. 



As in the previous experiments it was necessary to have 

 some means of judging if the rate of vaporization of the -alt 

 remained the same in the different experiments, the brightness 



of the streak was measured with an optical pyrometer, as 

 described in § 7. 



Since no direct comparison of the light emission and con- 

 ductivity seems to have been made, a brief series of experi- 

 ments were made on this point. The luminosity of the 

 coloured streak was varied by bringing different amounts of 

 salt into the flame, and was measured with the pyrometer, 

 the amount of energy radiated corresponding to the pyro- 

 meter readings being calculated as on p. 26. The con- 

 ductivity was measured with a cooled negative electrode 

 (see § 3), and the maximum current, i. e. the current 

 obtained when the luminous streak just touches the electrode, 

 in each case taken as the measure. 



Fig. 3 shows the relation obtained between radiated energy 



Fig. 3. 



Radiated energy — - 



ind conductivity 



Tl 



ey 



pro] 



tori ional, 



tie 



MM' 



asnrements air exhibited In. the Following table 



