256 



Sir James Dewar 



[Jan. 16, 



between 125° C. and 140° 0., and tinted with " scharlach-R." This, 

 though as light and mobile as the alcohol, gave no vapour which by 

 condensation might choke the narrow tubes of the thermoscope ; the 

 more viscous and dense sulphuric acid could then be dispensed with.) 

 Half to 1 minute was the usual period of exposure when comparing 

 fluxes of radiation, the displacements with the same aperture in 



DISPLACEMENT 

 CMS. 



IOCT RADIATION 



Z MINUTES 

 SCREENED RADIATION 



DISPLACEMENT 



CMS. 



IS" 



io- 



ta) UNSCREENED lOO* RAD" 

 ft) 100'CRAD"^ 



(O ir 



•025-m« MICA SCREEN 



O 1 2 3 *- 



Fig. 7. — Horizontal Scaled Tube. 



equaljjtimes being then compared ; when the inertia of the liquid 

 caused any initial distortion, the first 5 or 10 seconds were not 

 counted. A longer period was advisable when a very absorptive 

 screen gave low readings. The curve (Fig. 7 (2) ) shows this in the 

 case of mica, with which readings were continued for several minutes 

 without the occurrence of serious bending in the plotted results, as 

 the volume of gas expelled was still quite small. In this way the 



